Paper Digest: ICLR 2022 Highlights

February 8, 2022April 11, 2023 admin

Download ICLR-2022-Paper-Digests.pdf– Highlights of all ICLR-2022 papers (Oral, Spotlight, Poster). Readers can choose to read all these highlights on our console, which allows users to filter out papers using keywords and find related papers, patents, etc. To browse papers by author, here is a list of top ICLR-2022 authors. You may also like to explore our “Best Paper” Digest (ICLR), which lists the most influential ICLR papers in the last 4 years.

The International Conference on Learning Representations (ICLR) is one of the top machine learning conferences in the world. In 2022, it will be held online.

To help the community quickly catch up on the work presented in this conference, Paper Digest Team processed all accepted papers, and generated one highlight sentence (typically the main topic) for each paper. Readers are encouraged to read these machine generated highlights / summaries to quickly get the main idea of each paper.

Based in New York, Paper Digest is dedicated to helping people generate contents & reason over unstructured data. Different from black-box approaches, we build deep models on semantics, which allows results to be produced with explainations. Such models power this website, and are behind our services including “search engine”, “summarization”, “question answering”, and “literature review”.

If you do not want to miss interesting academic papers, you are welcome to sign up our daily paper digest service to get updates on new papers published in your area every day. You are also welcome to follow us on Twitter and Linkedin to get updated with new conference digests.

Paper Digest Team
New York City, New York, 10017
team@paperdigest.org

TABLE 1: Paper Digest: ICLR 2022 Highlights

	Paper	Author(s)
1	8-bit Optimizers Via Block-wise Quantization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We develop 8-bit optimizers reduce the memory footprint of training and maintain 32-bit optimizer performance across NLP/CV benchmarks.	Tim Dettmers; Mike Lewis; Sam Shleifer; Luke Zettlemoyer;
2	Ab-Initio Potential Energy Surfaces By Pairing GNNs with Neural Wave Functions Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a PESNet, a new network architecture that solves the Schr?dinger equation for multiple geometries simultaneously.	Nicholas Gao; Stephan G�nnemann;
3	A Biologically Interpretable Graph Convolutional Network to Link Genetic Risk Pathways and Imaging Phenotypes of Disease Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Biologically Informed Imaging-Genetics	Sayan Ghosal; Qiang Chen; Giulio Pergola; Aaron L Goldman; William Ulrich; Daniel R Weinberger; Archana Venkataraman;
4	Accelerated Policy Learning with Parallel Differentiable Simulation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose an efficient policy learning method leveraging the recent advance of differentiable simulation, and our method outperforms state-of-the-art algorithms in both sample efficiency and wall clock time on multiple challenging control tasks.	Jie Xu; Miles Macklin; Viktor Makoviychuk; Yashraj Narang; Animesh Garg; Fabio Ramos; Wojciech Matusik;
5	Acceleration of Federated Learning with Alleviated Forgetting in Local Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Here, we propose FedReg, an algorithm to accelerate FL with alleviated knowledge forgetting in the local training stage by regularizing locally trained parameters with the loss on generated pseudo data, which encode the knowledge of previous training data learned by the global model.	Chencheng Xu; Zhiwei Hong; Minlie Huang; Tao Jiang;
6	A Class of Short-term Recurrence Anderson Mixing Methods and Their Applications Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We develop a novel class of short-term recurrence Anderson mixing methods and validate its effectiveness in several applications including training neural networks.	Fuchao Wei; Chenglong Bao; Yang Liu;
7	A Comparison of Variable Selection Methods for Blockwise Diagonal Designs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A theoretical comparison of the Hamming errors for 6 different variable selection methods	Tracy Ke; Longlin Wang;
8	A Conditional Point Diffusion-Refinement Paradigm for 3D Point Cloud Completion Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To tackle this problem, we propose a novel Point Diffusion-Refinement (PDR) paradigm for point cloud completion.	Zhaoyang Lyu; Zhifeng Kong; Xudong XU; Liang Pan; Dahua Lin;
9	Active Hierarchical Exploration with Stable Subgoal Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a regularization to stabilize subgoal representation learning in goal-conditioned HRL and develop an active exploration strategy upon this stable representation.	Siyuan Li; Jin Zhang; Jianhao Wang; Yang Yu; Chongjie Zhang;
10	Actor-critic Is Implicitly Biased Towards High Entropy Optimal Policies Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that actor-critic, without any explicit exploration or regularization, can obtain an $\epsilon$-optimal high entropy policy in $\text{poly}(1/\epsilon)$ samples via a single trajectory without the usual uniform mixing assumptions.	Yuzheng Hu; Ziwei Ji; Matus Telgarsky;
11	Actor-Critic Policy Optimization in A Large-Scale Imperfect-Information Game Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A new actor-critic algorithm for approximating a Nash Equilibrium in the large-scale imperfect-information game 1v1 Mahjong.	Haobo Fu; Weiming Liu; Shuang Wu; Yijia Wang; Tao Yang; Kai Li; Junliang Xing; Bin Li; Bo Ma; QIANG FU; Yang Wei;
12	AdaAug: Learning Class- and Instance-adaptive Data Augmentation Policies Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel Automated Data Augmentation method called \texttt{AdaAug} to efficiently learn adaptive augmentation policies in a class-dependent and potentially instance-dependent manner.	Tsz-Him Cheung; Dit-Yan Yeung;
13	AdaMatch: A Unified Approach to Semi-Supervised Learning and Domain Adaptation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce AdaMatch, a unified solution that achieves state-of-the-art results for unsupervised domain adaptation (UDA), semi-supervised learning (SSL), and semi-supervised domain adaptation (SSDA).	David Berthelot; Rebecca Roelofs; Kihyuk Sohn; Nicholas Carlini; Alexey Kurakin;
14	Ada-NETS: Face Clustering Via Adaptive Neighbour Discovery in The Structure Space Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A novel algorithm named Ada-NETS is proposed to construct the clean graph for GCNs to cluster faces in this paper.	Yaohua Wang; Yaobin Zhang; Fangyi Zhang; Senzhang Wang; Ming Lin; YuQi Zhang; Xiuyu Sun;
15	Adaptive Control Flow in Transformers Improves Systematic Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We improve systematic generalization of Transformers on algorithmic tasks by introducing a novel attention mechanism and gating.	R�bert Csord�s; Kazuki Irie; J�rgen Schmidhuber;
16	AdaRL: What, Where, and How to Adapt in Transfer Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Efficient policy adaptation across domains by learning a parsimonious graphical representation that encodes changes in a compact way.	Biwei Huang; Fan Feng; Chaochao Lu; Sara Magliacane; Kun Zhang;
17	ADAVI: Automatic Dual Amortized Variational Inference Applied To Pyramidal Bayesian Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We automatically derive a variational family dual to a plate-enriched Hierarchical Bayesian Network and perform amortized inference.	Louis Rouillard; Demian Wassermann;
18	A Deep Variational Approach to Clustering Survival Data Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a novel semi-supervised probabilistic approach to cluster survival data	Laura Manduchi; Ricards Marcinkevics; Michela C. Massi; Thomas Weikert; Alexander Sauter; Verena Gotta; Timothy M�ller; Flavio Vasella; Marian C. Neidert; Marc Pfister; Bram Stieltjes; Julia E Vogt;
19	Adversarially Robust Conformal Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Multi-class calibration procedure that is provably robust to adversarial attacks	Asaf Gendler; Tsui-Wei Weng; Luca Daniel; Yaniv Romano;
20	Adversarial Retriever-Ranker for Dense Text Retrieval Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To address these challenges, we present Adversarial Retriever-Ranker (AR2), which consists of a dual-encoder retriever plus a cross-encoder ranker. We will make our code, models, and data publicly available.	Hang Zhang; Yeyun Gong; Yelong Shen; Jiancheng Lv; Nan Duan; Weizhu Chen;
21	Adversarial Robustness Through The Lens of Causality Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The first attempt towards using causality to understand and mitigate adversarial vulnerability.	Yonggang Zhang; Mingming Gong; Tongliang Liu; Gang Niu; Xinmei Tian; Bo Han; Bernhard Sch�lkopf; Kun Zhang;
22	Adversarial Support Alignment Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study the problem of aligning the supports of distributions.	Shangyuan Tong; Timur Garipov; Yang Zhang; Shiyu Chang; Tommi S. Jaakkola;
23	Adversarial Unlearning of Backdoors Via Implicit Hypergradient Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A minimax formulation of backdoor removal and an implicit gradient-based solver surpasses the state-of-art methods’ best results in higher efficacy, efficiency, robustness to variations in triggers, settings, poison ratio, and clean data size.	Yi Zeng; Si Chen; Won Park; Zhuoqing Mao; Ming Jin; Ruoxi Jia;
24	AEVA: Black-box Backdoor Detection Using Adversarial Extreme Value Analysis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Based on this observation, we propose the adversarial extreme value analysis(AEVA) algorithm to detect backdoors in black-box neural networks.	Junfeng Guo; Ang Li; Cong Liu;
25	A Fast and Accurate Splitting Method for Optimal Transport: Analysis and Implementation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We develop a fast and reliable method for solving large-scale optimal transport (OT) problems at an unprecedented combination of speed and accuracy.	Vien V. Mai; Jacob Lindb�ck; Mikael Johansson;
26	A Fine-Tuning Approach to Belief State Modeling Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A method for improving the accuracy of belief state models and for approximating public belief states at scale.	Samuel Sokota; Hengyuan Hu; David J Wu; J Zico Kolter; Jakob Nicolaus Foerster; Noam Brown;
27	A First-Occupancy Representation for Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce the first-occupancy representation, a modification of the successor representation which enables agents to perform rapid policy evaluation and planning for a class of ethologically important non-Markovian reward functions.	Ted Moskovitz; Spencer R Wilson; Maneesh Sahani;
28	A General Analysis of Example-Selection for Stochastic Gradient Descent Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we develop a broad condition on the sequence of examples used by SGD that is sufficient to prove tight convergence rates in both strongly convex and non-convex settings.	Yucheng Lu; Si Yi Meng; Christopher De Sa;
29	A Generalization of The Randomized Singular Value Decomposition Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The randomized SVD is generalized to multivariate Gaussian input vectors and Hilbert-Schmidt operators.	Nicolas Boulle; Alex Townsend;
30	A Generalized Weighted Optimization Method for Computational Learning and Inversion Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes a generalized weighted optimization method for computational learning and inversion with noisy data and derives generalization error bounds for various feature regression models, demonstrating better generalization capabilities.	Kui Ren; Yunan Yang; Bj�rn Engquist;
31	A Global Convergence Theory for Deep ReLU Implicit Networks Via Over-parameterization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: For a ReLU activated implicit neural network with infinitely many layers, we prove that randomly initialized gradient descent with a fixed step-size converges to a global minimum at a linear rate if the width m is the squared sample size n.	Tianxiang Gao; Hailiang Liu; Jia Liu; Hridesh Rajan; Hongyang Gao;
32	A Johnson-Lindenstrauss Framework for Randomly Initialized CNNs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study how the geometric representation of a dataset change after the application of each randomly initialized layer of a neural network.	Ido Nachum; Jan Hazla; Michael Gastpar; Anatoly Khina;
33	Almost Tight L0-norm Certified Robustness of Top-k Predictions Against Adversarial Perturbations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we derive the certified robustness against $\ell_0$-norm adversarial perturbation for top-$k$ prediction.	Jinyuan Jia; Binghui Wang; Xiaoyu Cao; Hongbin Liu; Neil Zhenqiang Gong;
34	A Loss Curvature Perspective on Training Instabilities of Deep Learning Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Our results suggest a unifying perspective on how disparate mitigation strategies for training instability ultimately address poor conditioning.	Justin Gilmer; Behrooz Ghorbani; Ankush Garg; Sneha Kudugunta; Behnam Neyshabur; David Cardoze; George Edward Dahl; Zachary Nado; Orhan Firat;
35	AlphaZero-based Proof Cost Network to Aid Game Solving Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes a novel approach to solving problems by modifying the training target of the AlphaZero algorithm, such that it prioritizes solving the game quickly, rather than winning.	Ti-Rong Wu; Chung-Chin Shih; Ting Han Wei; Meng-Yu Tsai; Wei-Yuan Hsu; I-Chen Wu;
36	Amortized Implicit Differentiation for Stochastic Bilevel Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide a unified framework for analyzing bilevel optimization algorithm based on approximate implicit differentiation with a warm-start strategy	Michael Arbel; Julien Mairal;
37	Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a model that address synthesis planning and synthesizable molecular design simultaneously.	Wenhao Gao; Roc�o Mercado; Connor W. Coley;
38	An Agnostic Approach to Federated Learning with Class Imbalance Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper we propose a novel agnostic constrained learning formulation to tackle the class imbalance problem in FL, without requiring further information beyond the standard FL objective.	Zebang Shen; Juan Cervino; Hamed Hassani; Alejandro Ribeiro;
39	Analytic-DPM: An Analytic Estimate of The Optimal Reverse Variance in Diffusion Probabilistic Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose an analytic framework of estimating the optimal reverse variance in DPMs.	Fan Bao; Chongxuan Li; Jun Zhu; Bo Zhang;
40	Analyzing and Improving The Optimization Landscape of Noise-Contrastive Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This work theoretically explains the difficulty of optimizing the NCE loss when the noise distribution is poor, and provides a provably efficient solution consisting of normalized gradient descent (NGD) combined with the proposed \emph{eNCE} loss.	Bingbin Liu; Elan Rosenfeld; Pradeep Kumar Ravikumar; Andrej Risteski;
41	An Autoregressive Flow Model for 3D Molecular Geometry Generation from Scratch Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a novel method for 3D molecular geometry generation from scratch.	Youzhi Luo; Shuiwang Ji;
42	Ancestral Protein Sequence Reconstruction Using A Tree-structured Ornstein-Uhlenbeck Variational Autoencoder Related Papers Related Patents Related Grants Related Orgs Related Experts View Abstract: We introduce a deep generative model for representation learning of biological sequences that, unlike existing models, explicitly represents the evolutionary process. The model …	Lys Sanz Moreta; Ola R�nning; Ahmad Salim Al-Sibahi; Jotun Hein; Douglas Theobald; Thomas Hamelryck;
43	A Neural Tangent Kernel Perspective of Infinite Tree Ensembles Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: By considering an ensemble of infinite trees, we introduce and study the Tree Neural Tangent Kernel (TNTK), which provides new insights into the behavior of the infinite ensemble of soft trees.	Ryuichi Kanoh; Mahito Sugiyama;
44	An Experimental Design Perspective on Exploration in Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We draw a connection between Bayesian Optimal Experiment Design and RL to develop an acquisition function to guide data collection in model based RL leading to improved sample efficiency.	Viraj Mehta; Biswajit Paria; Jeff Schneider; Willie Neiswanger; Stefano Ermon;
45	An Explanation of In-context Learning As Implicit Bayesian Inference Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In-context learning emerges both theoretically and empirically when the pretraining distribution is a mixture distribution, resulting in the language model implicitly performing Bayesian inference in its forward pass.	Sang Michael Xie; Aditi Raghunathan; Percy Liang; Tengyu Ma;
46	An Information Fusion Approach to Learning with Instance-Dependent Label Noise Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This work is the first time to realize and bridge the gap between clean and noisy empirical distribution mismatch.	Zhimeng Jiang; Kaixiong Zhou; Zirui Liu; Li Li; Rui Chen; Soo-Hyun Choi; Xia Hu;
47	Anisotropic Random Feature Regression in High Dimensions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We derive exact asymptotic formulas for the total error, bias, and variance of random feature regression with anisotropic inputs and target weights, and identify a new type of singularity in sample-wise learning curves.	Gabriel Mel; Jeffrey Pennington;
48	Anomaly Detection for Tabular Data with Internal Contrastive Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: An anomaly detection method based on the ability to predict the masked out part in a vector.	Tom Shenkar; Lior Wolf;
49	Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper detects time series anomalies from a new association-based dimension. We find an inherently normal-abnormal distinguishable evidence as Association Discrepancy. Co-designed with this evidence, our model achieves the SOTA on six benchmarks.	Jiehui Xu; Haixu Wu; Jianmin Wang; Mingsheng Long;
50	An Operator Theoretic View On Pruning Deep Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Koopman operator theoretic methods are extended to deep neural network pruning and are shown to provide new insight into existing work.	William T Redman; MARIA FONOBEROVA; Ryan Mohr; Yannis Kevrekidis; Igor Mezic;
51	Anti-Concentrated Confidence Bonuses For Scalable Exploration Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce anti-concentrated confidence bounds for efficiently approximating the elliptical bonus, using an ensemble of regressors trained to predict random noise from policy network-derived features.	Jordan T. Ash; Cyril Zhang; Surbhi Goel; Akshay Krishnamurthy; Sham M. Kakade;
52	An Unconstrained Layer-Peeled Perspective on Neural Collapse Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We investigate how the gradient flow converges to a neural collapse solution in an unconstrained model.	Wenlong Ji; Yiping Lu; Yiliang Zhang; Zhun Deng; Weijie J Su;
53	Approximation and Learning with Deep Convolutional Models: A Kernel Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study the inductive bias of multi-layer convolutional models through a kernel lens, showing generalization benefits of various architectural choices such as locality, depth, and pooling layers.	Alberto Bietti;
54	A Program to Build E(N)-Equivariant Steerable CNNs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We derive a general method to build G-steerable kernel spaces for equivariant steerable CNNs	Gabriele Cesa; Leon Lang; Maurice Weiler;
55	A Reduction-Based Framework for Conservative Bandits and Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We give general framework that turns upper and lower bounds in non-conservative settings to bounds in conservative settings.	Yunchang Yang; Tianhao Wu; Han Zhong; Evrard Garcelon; Matteo Pirotta; Alessandro Lazaric; Liwei Wang; Simon Shaolei Du;
56	A Relational Intervention Approach for Unsupervised Dynamics Generalization in Model-Based Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper proposes a new model-based RL that could generalize to new environments.	Jiaxian Guo; Mingming Gong; Dacheng Tao;
57	ARTEMIS: Attention-based Retrieval with Text-Explicit Matching and Implicit Similarity Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Our work aims at shedding new light on the task by looking at it through the prism of two familiar and related frameworks: text-to-image and image-to-image retrieval.	Ginger Delmas; Rafael S. Rezende; Gabriela Csurka; Diane Larlus;
58	AS-MLP: An Axial Shifted MLP Architecture for Vision Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We design the first MLP-based architecture for downstream tasks. It achieves competitive performance compared to the transformer-based architecture, which establishes a new strong baseline of MLP-based architecture.	Dongze Lian; Zehao Yu; Xing Sun; Shenghua Gao;
59	Assessing Generalization of SGD Via Disagreement Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide a surprisingly simple technique to accurately estimate the test error of deep neural networks using unlabeled data and we prove that this works because SGD ensembles are naturally well-calibrated.	Yiding Jiang; Vaishnavh Nagarajan; Christina Baek; J Zico Kolter;
60	Associated Learning: An Alternative to End-to-End Backpropagation That Works on CNN, RNN, and Transformer Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper studies Associate Learning, an alternative methodology to the end-to-end backpropagation	Dennis Y.H. Wu; Dinan Lin; Vincent Chen; Hung-Hsuan Chen;
61	A Statistical Framework for Efficient Out of Distribution Detection in Deep Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We frame Out Of Distribution (OOD) detection in DNNs as a statistical hypothesis testing problem.	Matan Haroush; Tzviel Frostig; Ruth Heller; Daniel Soudry;
62	A Tale of Two Flows: Cooperative Learning of Langevin Flow and Normalizing Flow Toward Energy-Based Model Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Joint learning of a short-run MCMC generator and a normalizing flow in the context of energy-based model for image representation and generation.	Jianwen Xie; Yaxuan Zhu; Jun Li; Ping Li;
63	A Theoretical Analysis on Feature Learning in Neural Networks: Emergence from Inputs and Advantage Over Fixed Features Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Theoretical analysis of feature learning in neural networks on practically motivated learning problems, showing it strongly depends on the structure of the input distribution and leads to the advantage over fixed feature methods like kernels.	Zhenmei Shi; Junyi Wei; Yingyu Liang;
64	A Theory of Tournament Representations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We develop a theory to understand tournament representations i.e. structurally characterise when a tournament graph can be represented in lower dimensions using a skew symmetric matrix.	Arun Rajkumar; Vishnu Veerathu; Abdul Bakey Mir;
65	Attention-based Interpretability with Concept Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We the Concept Transformer an architecture that generalization attention from low-level input features to high-level concepts as a mechanism to ensure the interpretability of attention scores within a given application domain.	Mattia Rigotti; Christoph Miksovic; Ioana Giurgiu; Thomas Gschwind; Paolo Scotton;
66	Audio Lottery: Speech Recognition Made Ultra-Lightweight, Noise-Robust, and Transferable Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We for the first time investigate three unique properties that were rarely studied in previous LTH research but are key to user-interactive ASR devices, bringing new insights to both LTH theory and lightweight ASR research.	Shaojin Ding; Tianlong Chen; Zhangyang Wang;
67	Augmented Sliced Wasserstein Distances Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we propose a new family of distance metrics, called augmented sliced Wasserstein distances (ASWDs), constructed by first mapping samples to higher-dimensional hypersurfaces parameterized by neural networks.	Xiongjie Chen; Yongxin Yang; Yunpeng Li;
68	A Unified Contrastive Energy-based Model for Understanding The Generative Ability of Adversarial Training Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we demystify this phenomenon by developing a unified probabilistic framework, called Contrastive Energy-based Models (CEM).	Yifei Wang; Yisen Wang; Jiansheng Yang; Zhouchen Lin;
69	A Unified Wasserstein Distributional Robustness Framework for Adversarial Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Abstract: It is well-known that deep neural networks (DNNs) are susceptible to adversarial attacks, exposing a severe fragility of deep learning systems. As the result, adversarial training …	Anh Tuan Bui; Trung Le; Quan Hung Tran; He Zhao; Dinh Phung;
70	Automated Self-Supervised Learning for Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: An automated self-supervised learning algorithm for graph neural networks.	Wei Jin; Xiaorui Liu; Xiangyu Zhao; Yao Ma; Neil Shah; Jiliang Tang;
71	Automatic Loss Function Search for Predict-Then-Optimize Problems with Strong Ranking Property Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To properly connect the prediction loss with the optimization goal, in this paper we propose a total group preorder (TGP) loss and its differential version called approximated total group preorder (ATGP) loss for predict-then-optimize (PTO) problems with strong ranking property.	Boshi Wang; Jialin Yi; Hang Dong; Bo Qiao; Chuan Luo; Qingwei Lin;
72	Autonomous Learning of Object-Centric Abstractions for High-Level Planning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We learn object-centric PDDL representations directly from raw observation data	Steven James; Benjamin Rosman; George Konidaris;
73	Autonomous Reinforcement Learning: Formalism and Benchmarking Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we aim to address this discrepancy by laying out a framework for Autonomous Reinforcement Learning (ARL): reinforcement learning where the agent not only learns through its own experience, but also contends with lack of human supervision to reset between trials.	Archit Sharma; Kelvin Xu; Nikhil Sardana; Abhishek Gupta; Karol Hausman; Sergey Levine; Chelsea Finn;
74	Autoregressive Quantile Flows for Predictive Uncertainty Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Using Quantile Flows for Predictive and Generative Data Modeling and Generation	Phillip Si; Volodymyr Kuleshov; Allan Bishop;
75	Auto-scaling Vision Transformers Without Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We automate the design and scaling of vision transformers without any training, achieving state-of-the-art performance on ImageNet classification and COCO object detection.	Wuyang Chen; Wei Huang; Xianzhi Du; Xiaodan Song; Zhangyang Wang; Denny Zhou;
76	Auto-Transfer: Learning to Route Transferable Representations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper offers a novel transfer method that uses a routing function to select source-target pairs as well as aggregation functions that combine source and target features (as opposed to matching source to target features).	Keerthiram Murugesan; Vijay Sadashivaiah; Ronny Luss; Karthikeyan Shanmugam; Pin-Yu Chen; Amit Dhurandhar;
77	Axiomatic Explanations for Visual Search, Retrieval, and Similarity Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that cooperative game theory provides an axiomatic characterization of model interpretability for visual search, retrieval, and similarity learning architectures	Mark Hamilton; Scott Lundberg; Stephanie Fu; Lei Zhang; William T. Freeman;
78	A Zest of LIME: Towards Architecture-Independent Model Distances Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose an architecture-independent distance metric that measures the similarity between ML models by comparing their global behaviors, approximated using LIME.	Hengrui Jia; Hongyu Chen; Jonas Guan; Ali Shahin Shamsabadi; Nicolas Papernot;
79	Back2Future: Leveraging Backfill Dynamics for Improving Real-time Predictions in Future Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study the problem of multi-variate backfill for both features and targets and show how to leverage our insights for more general neural framework to improve both model predictions and evaluation We construct a detailed dataset composed of relevant signals over the past year of the pandemic.	Harshavardhan Kamarthi; Alexander Rodr�guez; B. Aditya Prakash;
80	Backdoor Defense Via Decoupling The Training Process Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We reveal that the hidden backdoors are embedded in the feature space mostly due to the end-to-end supervised training paradigm, based on which we propose a simple yet effective decoupling-based training method for backdoor defense.	Kunzhe Huang; Yiming Li; Baoyuan Wu; Zhan Qin; Kui Ren;
81	BadPre: Task-agnostic Backdoor Attacks to Pre-trained NLP Foundation Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we propose BadPre, the first backdoor attack against various downstream models built based on pre-trained NLP models.	Kangjie Chen; Yuxian Meng; Xiaofei Sun; Shangwei Guo; Tianwei Zhang; Jiwei Li; Chun Fan;
82	Bag of Instances Aggregation Boosts Self-supervised Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper proposes a new self-supervised distillation method which aggregates related instances bagged by the teacher and shows stronger performance than previous relation-agnostic methods.	Haohang Xu; Jiemin Fang; XIAOPENG ZHANG; Lingxi Xie; Xinggang Wang; Wenrui Dai; Hongkai Xiong; Qi Tian;
83	BAM: Bayes with Adaptive Memory Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We augment Bayes with memory to generalize many frameworks and overcome limitations of traditional methods in non-stationary settings	Josue Nassar; Jennifer Rogers Brennan; Ben Evans; Kendall Lowrey;
84	Bandit Learning with Joint Effect of Incentivized Sampling, Delayed Sampling Feedback, and Self-Reinforcing User Preferences Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we consider a new multi-armed bandit (MAB) framework motivated by three common complications in online recommender systems in practice: (i) the platform (learning agent) cannot sample an intended product directly and has to incentivize customers to select this product (e.g., promotions and coupons); (ii) customer feedbacks are often received later than their selection times; and (iii) customer preferences among products are influenced and reinforced by historical feedbacks.	Tianchen Zhou; Jia Liu; Chaosheng Dong; Yi Sun;
85	Bayesian Framework for Gradient Leakage Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a theoretical framework for analysis of the Bayes optimal adversary for gradient leakage, and perform evaluation of existing defenses.	Mislav Balunovic; Dimitar Iliev Dimitrov; Robin Staab; Martin Vechev;
86	Bayesian Modeling and Uncertainty Quantification for Learning to Optimize: What, Why, and How Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Our study represents the first effort to recognize and quantify the uncertainty of the optimization algorithm.	Yuning You; Yue Cao; Tianlong Chen; Zhangyang Wang; Yang Shen;
87	Bayesian Neural Network Priors Revisited Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Using BNN priors that are not isotropic Gaussians can improve performance and reduce the cold posterior effect.	Vincent Fortuin; Adri� Garriga-Alonso; Sebastian W. Ober; Florian Wenzel; Gunnar Ratsch; Richard E Turner; Mark van der Wilk; Laurence Aitchison;
88	Benchmarking The Spectrum of Agent Capabilities Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce Crafter, an open world survival game with visual inputs that evaluates a wide range of general abilities within a single environment.	Danijar Hafner;
89	$\beta$-Intact-VAE: Identifying and Estimating Causal Effects Under Limited Overlap Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: See all these naturally in one: limited overlap, prognostic score, identifiable VAE, balanced representation Learning, CATE error bounds.	Pengzhou Abel Wu; Kenji Fukumizu;
90	Better Supervisory Signals By Observing Learning Paths Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A study of how and why teachers in knowledge distillation end up with better supervisory signals than the original labels.	Yi Ren; Shangmin Guo; Danica J. Sutherland;
91	Bi-linear Value Networks for Multi-goal Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a bilinear value function for multi-goal reinforcement learning and show superior sample efficiency and generalizability.	Ge Yang; Zhang-Wei Hong; Pulkit Agrawal;
92	Blaschke Product Neural Networks (BPNN): A Physics-Infused Neural Network for Phase Retrieval of Meromorphic Functions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To this end, we propose a physics-infused deep neural network based on the Blaschke products for phase retrieval.	Juncheng Dong; Simiao Ren; Yang Deng; Omar Khatib; Jordan Malof; Mohammadreza Soltani; Willie Padilla; Vahid Tarokh;
93	Boosted Curriculum Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A novel approach for curriculum RL that increases the representativeness of the functional space as new, increasingly complex, tasks from the curriculum are presented to the agent.	Pascal Klink; Carlo D’Eramo; Jan Peters; Joni Pajarinen;
94	Boosting Randomized Smoothing with Variance Reduced Classifiers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show — theoretically and empirically — that ensembles reduce variance under randomized smoothing, yielding higher certified accuracy, leading to a new state-of-the-art on CIFAR-10 and ImageNet.	Mikl�s Z. Horv�th; Mark Niklas Mueller; Marc Fischer; Martin Vechev;
95	Bootstrapping Semantic Segmentation with Regional Contrast Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a pixel-level contrastive learning framework to achieve a high-quality semantic segmeantation model trained with very few human annotations.	Shikun Liu; Shuaifeng Zhi; Edward Johns; Andrew Davison;
96	Bregman Gradient Policy Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we design a novel Bregman gradient policy optimization framework for reinforcement learning based on Bregman divergences and momentum techniques.	Feihu Huang; Shangqian Gao; Heng Huang;
97	Bridging Recommendation and Marketing Via Recurrent Intensity Modeling Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We repurpose an item-recommender system to recommend users for item providers for the purpose of content promotion and diversity.	Yifei Ma; Ge Liu; Anoop Deoras;
98	Bridging The Gap: Providing Post-Hoc Symbolic Explanations for Sequential Decision-Making Problems with Inscrutable Representations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper introduces methods for providing contrastive explanations in terms of user-specified concepts for sequential decision-making settings where the system’s model of the task may be best represented as an inscrutable model.	Sarath Sreedharan; Utkarsh Soni; Mudit Verma; Siddharth Srivastava; Subbarao Kambhampati;
99	Bundle Networks: Fiber Bundles, Local Trivializations, and A Generative Approach to Exploring Many-to-one Maps Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We draw from the theory of fiber bundles in (differential) topology to create a principled approach to generative models that allow us to learn and sample from the fiber over a point in a many-to-one map.	Nico Courts; Henry Kvinge;
100	Byzantine-Robust Learning on Heterogeneous Datasets Via Bucketing Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Byzantine-robust distributed learning with heterogeneous data distribution	Sai Praneeth Karimireddy; Lie He; Martin Jaggi;
101	CADDA: Class-wise Automatic Differentiable Data Augmentation for EEG Signals Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new gradient-based automatic data augmentation technique where samples are transformed based on their labels, and demonstrate its relevance on EEG sleep staging data.	C�dric Rommel; Thomas Moreau; Joseph Paillard; Alexandre Gramfort;
102	Can An Image Classifier Suffice For Action Recognition? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the idea of super images to re-purpose an image classifer for action recognition.	Quanfu Fan; Chun-Fu Chen; Rameswar Panda;
103	Capacity of Group-invariant Linear Readouts from Equivariant Representations: How Many Objects Can Be Linearly Classified Under All Possible Views? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We compute the fraction of linearly separable dichotomies of representations formed by group actions and apply these results to intermediate representations of convolutional neural networks.	Matthew Farrell; Blake Bordelon; Shubhendu Trivedi; Cengiz Pehlevan;
104	Case-based Reasoning for Better Generalization in Text-Adventure Games Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In a departure from deep RL approaches, in this paper, we propose a general method inspired by case-based reasoning to train agents and generalize out of the training distribution.	Mattia Atzeni; Shehzaad Zuzar Dhuliawala; Keerthiram Murugesan; MRINMAYA SACHAN;
105	Causal Contextual Bandits with Targeted Interventions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A new, more realistic, formalism of contextual bandits involving causal side-information and targeted interventions, along with a novel algorithm that exploits features of the new setting such as information leakage to learn good policies quickly.	Chandrasekar Subramanian; Balaraman Ravindran;
106	Causal ImageNet: How to Discover Spurious Features in Deep Learning? Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A scalable framework for discovering spurious features of deep neural networks Using this methodology, we introduce the {\it Causal Imagenet} dataset containing causal and spurious masks for a large set of samples from Imagenet.	Sahil Singla; Soheil Feizi;
107	CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Along with the pseudo labels, a weight-sharing triple-branch transformer framework is proposed to apply self-attention and cross-attention for source/target feature learning and source-target domain alignment, respectively.	Tongkun Xu; Weihua Chen; Pichao WANG; Fan Wang; Hao Li; Rong Jin;
108	Certified Robustness for Deep Equilibrium Models Via Interval Bound Propagation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To develop certifiably robust deep equilibrium (DEQ) models, we propose the IBP-MonDEQ layer, a DEQ layer where interval bounds on the output can be obtained by solving an additional fixed-point equation inspired by interval bound propagation.	Colin Wei; J Zico Kolter;
109	Chaos Is A Ladder: A New Understanding of Contrastive Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose a new guarantee on the downstream performance without resort to the conditional independence assumption that is widely adopted in previous work but hardly holds in practice.	Yifei Wang; Qi Zhang; Yisen Wang; Jiansheng Yang; Zhouchen Lin;
110	Charformer: Fast Character Transformers Via Gradient-based Subword Tokenization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Fast Token-Free Models	Yi Tay; Vinh Q. Tran; Sebastian Ruder; Jai Gupta; Hyung Won Chung; Dara Bahri; Zhen Qin; Simon Baumgartner; Cong Yu; Donald Metzler;
111	Chemical-Reaction-Aware Molecule Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We make use of chemical reactions to improve the generalization ability of learned molecule embeddings	Hongwei Wang; Weijiang Li; Xiaomeng Jin; Kyunghyun Cho; Heng Ji; Jiawei Han; Martin Burke;
112	Chunked Autoregressive GAN for Conditional Waveform Synthesis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We improve the state-of-the-art of conditional waveform synthesis by combining the strengths of GANs and autoregression	Max Morrison; Rithesh Kumar; Kundan Kumar; Prem Seetharaman; Aaron Courville; Yoshua Bengio;
113	Churn Reduction Via Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show distillation is a principled and practical solution to churn reduction.	Heinrich Jiang; Harikrishna Narasimhan; Dara Bahri; Andrew Cotter; Afshin Rostamizadeh;
114	CKConv: Continuous Kernel Convolution For Sequential Data Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We provide a continuous parameterization to convolutional kernels, with which several advantages upon conventional (discrete) parameterizations are obtained.	David W. Romero; Anna Kuzina; Erik J Bekkers; Jakub Mikolaj Tomczak; Mark Hoogendoorn;
115	Clean Images Are Hard to Reblur: Exploiting The Ill-Posed Inverse Task for Dynamic Scene Deblurring Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Reblurring, the inverse task of deblurring, is used for supervised/self-supervised learning of deblurring and improves the image sharpness.	Seungjun Nah; Sanghyun Son; Jaerin Lee; Kyoung Mu Lee;
116	ClimateGAN: Raising Climate Change Awareness By Generating Images of Floods Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper presents a model to robustly produce photo-realistic images of floods for raising climate change awareness, leveraging unsupervised domain adaptation and conditional image generation.	Victor Schmidt; Alexandra Luccioni; M�lisande Teng; Tianyu Zhang; Alexia Reynaud; Sunand Raghupathi; Gautier Cosne; Adrien Juraver; Vahe Vardanyan; Alex Hern�ndez-Garc�a; Yoshua Bengio;
117	CoBERL: Contrastive BERT for Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A new loss and an improved architecture to efficiently train attentional models in reinforcement learning.	Andrea Banino; Adria Puigdomenech Badia; Jacob C Walker; Tim Scholtes; Jovana Mitrovic; Charles Blundell;
118	CodeTrek: Flexible Modeling of Code Using An Extensible Relational Representation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a relational database representation and corresponding neural module for source code and show its potential on program understanding tasks	Pardis Pashakhanloo; Aaditya Naik; Yuepeng Wang; Hanjun Dai; Petros Maniatis; Mayur Naik;
119	Cold Brew: Distilling Graph Node Representations with Incomplete or Missing Neighborhoods Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Improve strict cold start performances for graph minings with a knowledge distillation framework.	Wenqing Zheng; Edward W Huang; Nikhil Rao; Sumeet Katariya; Zhangyang Wang; Karthik Subbian;
120	Collapse By Conditioning: Training Class-conditional GANs with Limited Data Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Motivated by this observation, we propose a training strategy for conditional GANs (cGANs) that effectively prevents the observed mode-collapse by leveraging unconditional learning.	Mohamad Shahbazi; Martin Danelljan; Danda Pani Paudel; Luc Van Gool;
121	Communication-Efficient Actor-Critic Methods for Homogeneous Markov Games Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we formally characterize a subclass of cooperative Markov games where agents exhibit a certain form of homogeneity such that policy sharing provably incurs no suboptimality.	Dingyang Chen; Yile Li; Qi Zhang;
122	ComPhy: Compositional Physical Reasoning of Objects and Events from Videos Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce a new dataset for Compositional Physical Reasoning	Zhenfang Chen; Kexin Yi; Yunzhu Li; Mingyu Ding; Antonio Torralba; Joshua B. Tenenbaum; Chuang Gan;
123	Complete Verification Via Multi-Neuron Relaxation Guided Branch-and-Bound Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We obtain a state-of-the-art GPU-based neural network verifier by leveraging tight multi-neuron constraints in a Branch-and-Bound setting.	Claudio Ferrari; Mark Niklas Mueller; Nikola Jovanovic; Martin Vechev;
124	Compositional Attention: Disentangling Search and Retrieval Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Recombining search and retrieval mechanisms of multi-head attention in a disentangled and flexible manner for better representational capacity and generalization.	Sarthak Mittal; Sharath Chandra Raparthy; Irina Rish; Yoshua Bengio; Guillaume Lajoie;
125	Compositional Training for End-to-End Deep AUC Maximization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel end-to-end training framework with a provable stochastic algorithm for deep AUC maximization.	Zhuoning Yuan; Zhishuai Guo; Nitesh Chawla; Tianbao Yang;
126	CoMPS: Continual Meta Policy Search Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Cotinual meta-reinforcement learning accelerates task learning, via repeated meta off-policy search.	Glen Berseth; Zhiwei Zhang; Grace Zhang; Chelsea Finn; Sergey Levine;
127	Concurrent Adversarial Learning for Large-Batch Training Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose to use adversarial learning to increase the batch size in large-batch training.	Yong Liu; Xiangning Chen; Minhao Cheng; Cho-Jui Hsieh; Yang You;
128	Conditional Contrastive Learning with Kernel Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper presents Conditional Contrastive Learning with Kernel (CCL-K) for conditional contrastive learning tasks under the scenario when we have insufficient data for some values of the condioning variable.	Yao-Hung Hubert Tsai; Tianqin Li; Martin Q. Ma; Han Zhao; Kun Zhang; Louis-Philippe Morency; Ruslan Salakhutdinov;
129	Conditional Image Generation By Conditioning Variational Auto-Encoders Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We create fast-to-train conditional VAEs using amortized inference in pretrained unconditional VAEs, and demonstrate diverse samples on image completion tasks.	William Harvey; Saeid Naderiparizi; Frank Wood;
130	Conditional Object-Centric Learning from Video Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we instead take a weakly-supervised approach and focus on how 1) using the temporal dynamics of video data in the form of optical flow and 2) conditioning the model on simple object location cues can be used to enable segmenting and tracking objects in significantly more realistic synthetic data.	Thomas Kipf; Gamaleldin Fathy Elsayed; Aravindh Mahendran; Austin Stone; Sara Sabour; Georg Heigold; Rico Jonschkowski; Alexey Dosovitskiy; Klaus Greff;
131	Conditioning Sequence-to-sequence Networks with Learned Activations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Conditioning neural networks by learning the layer activations based on the conditioning vector	Alberto Gil Couto Pimentel Ramos; Abhinav Mehrotra; Nicholas Donald Lane; Sourav Bhattacharya;
132	ConFeSS: A Framework for Single Source Cross-Domain Few-Shot Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose a framework for few-shot learning coined as ConFeSS (Contrastive Learning and Feature Selection System) that tackles large domain shift between base and novel categories.	Debasmit Das; Sungrack Yun; Fatih Porikli;
133	Confidence Adaptive Anytime Pixel-Level Recognition Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: First single-model anytime approach for pixel-level visual recognition; our model with redesigned exits and confidence adaptivity enables anytime inference, achieves the same level of final accuracy, and significantly reduces total computation.	Zhuang Liu; Hung-Ju Wang; Zhiqiu Xu; Trevor Darrell; Evan Shelhamer;
134	Consistent Counterfactuals for Deep Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Counterfactual explanations are often inconsistent between virtually identical deep models. We introduce a new method to increase consistency while keeping costs low relative to other fixes.	Emily Black; Zifan Wang; Matt Fredrikson;
135	Constrained Graph Mechanics Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose Graph Mechanics Network (GMN) which is combinatorially efficient, equivariant and constraint-aware.	Wenbing Huang; Jiaqi Han; Yu Rong; Tingyang Xu; Fuchun Sun; Junzhou Huang;
136	Constrained Physical-Statistics Models for Dynamical System Identification and Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose to incorporate constraints in the learning of hybrid physical and data driven dynamical models.	J�r�mie DONA; Marie D�chelle; patrick gallinari; Marina Levy;
137	Constrained Policy Optimization Via Bayesian World Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Solving constrained Markov decision processes with Bayesian model-based reinforcement learning.	Yarden As; Ilnura Usmanova; Sebastian Curi; Andreas Krause;
138	Constraining Linear-chain CRFs to Regular Languages Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: CRFs can be efficiently constrained to arbitrary regular languages, enforcing nonlocal constraints on their outputs.	Sean Papay; Roman Klinger; Sebastian Pado;
139	Constructing A Good Behavior Basis for Transfer Using Generalized Policy Updates Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Specifically, we consider the framework of generalized policy evaluation and improvement, in which the rewards for all tasks of interest are assumed to be expressible as a linear combination of a fixed set of features.	Safa Alver; Doina Precup;
140	Constructing Orthogonal Convolutions in An Explicit Manner Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we exploit the relation between the singular values of the convolution layer’s Jacobian and the structure of the convolution kernel.	Tan Yu; Jun Li; YUNFENG CAI; Ping Li;
141	Contact Points Discovery for Soft-Body Manipulations with Differentiable Physics Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a contact pose discovery method that guides the stand-alone differentiable physics solver to complete various soft-body manipulation tasks.	Sizhe Li; Zhiao Huang; Tao Du; Hao Su; Joshua B. Tenenbaum; Chuang Gan;
142	Context-Aware Sparse Deep Coordination Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel method for learning sparse coordination graphs that can be theoretically justified and can significantly reduce communication overhead and improve learning performance of deep coordination graphs. To empirically evaluate our method, we present the Multi-Agent COordination (MACO) benchmark by collecting classic coordination problems in the literature, increasing their difficulty, and classifying them into different types.	Tonghan Wang; Liang Zeng; Weijun Dong; Qianlan Yang; Yang Yu; Chongjie Zhang;
143	Contextualized Scene Imagination for Generative Commonsense Reasoning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work aims at tackling generative commonsense reasoning by allowing machines to imagine a reasonable scene before generating text.	PeiFeng Wang; Jonathan Zamora; Junfeng Liu; Filip Ilievski; Muhao Chen; Xiang Ren;
144	Continual Learning with Filter Atom Swapping Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we first enforce a low-rank filter subspace by decomposing convolutional filters within each network layer over a small set of filter atoms.	Zichen Miao; Ze Wang; Wei Chen; Qiang Qiu;
145	Continual Learning with Recursive Gradient Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes a novel method for continual learning in a fixed capacity network in the non-replay regime, which minimizes the loss on the current task while also minimizing an upper bound of loss increment on previous tasks.	Hao Liu; Huaping Liu;
146	Continual Normalization: Rethinking Batch Normalization for Online Continual Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A negative effect of BN in online continual learning and a simple strategy to alleviate it.	Quang Pham; Chenghao Liu; Steven HOI;
147	Continuously Discovering Novel Strategies Via Reward-Switching Policy Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Reward-Switching Policy Optimization (RSPO), a paradigm to discover diverse strategies in complex RL environments by iteratively finding novel policies that are both locally optimal and sufficiently different from existing ones.	Zihan Zhou; Wei Fu; Bingliang Zhang; Yi Wu;
148	Continuous-Time Meta-Learning with Forward Mode Differentiation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: COMLN is a new meta-learning algorithm, where adaptation follows a gradient flow. It enables learning the amount of adaptation using SGD. We devise a novel efficient algorithm to compute the meta-gradients of COMLN, based on forward-mode diff.	Tristan Deleu; David Kanaa; Leo Feng; Giancarlo Kerg; Yoshua Bengio; Guillaume Lajoie; Pierre-Luc Bacon;
149	Contrastive Clustering to Mine Pseudo Parallel Data for Unsupervised Translation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a fine-tuning loss that enables pre-trained model’s ability to mine pseudo-parallel data for fully unsupervised machine translation.	Xuan-Phi Nguyen; Hongyu Gong; Yun Tang; Changhan Wang; Philipp Koehn; Shafiq Joty;
150	Contrastive Fine-grained Class Clustering Via Generative Adversarial Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We proposed a method for unsupervised fine-grained class clustering that leverages the information-theoretic regularization term based on contrastive loss.	Yunji Kim; Jung-Woo Ha;
151	Contrastive Learning Is Just Meta-Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We discuss the close relationship between contrastive learning and meta-learning, and we propose a meta-learning framework for self-supervised learning (SSL) along with meta-specific methods to improve contrastive learning performance for SSL.	Renkun Ni; Manli Shu; Hossein Souri; Micah Goldblum; Tom Goldstein;
152	Controlling Directions Orthogonal to A Classifier Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We develop a notion of orthogonality in classifier, and the corresponding construction and utility.	Yilun Xu; Hao He; Tianxiao Shen; Tommi S. Jaakkola;
153	Controlling The Complexity and Lipschitz Constant Improves Polynomial Nets Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide sample complexity results and bounds on the Lipschitz constant of polynomial networks, which we use to construct a regularization scheme that improves the robustness against adversarial noise.	Zhenyu Zhu; Fabian Latorre; Grigorios Chrysos; Volkan Cevher;
154	Convergent and Efficient Deep Q Learning Algorithm Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we show that DQN can indeed diverge and cease to operate in realistic settings.	Zhikang T. Wang; Masahito Ueda;
155	Convergent Boosted Smoothing for Modeling GraphData with Tabular Node Features Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We develop a convergent method for combining boosting and graph propagation layers.	Jiuhai Chen; Jonas Mueller; Vassilis N. Ioannidis; Soji Adeshina; Yangkun Wang; Tom Goldstein; David Wipf;
156	Coordination Among Neural Modules Through A Shared Global Workspace Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: communication among different specialist using a shared workspace allowing higher order interactions	Anirudh Goyal; Aniket Rajiv Didolkar; Alex Lamb; Kartikeya Badola; Nan Rosemary Ke; Nasim Rahaman; Jonathan Binas; Charles Blundell; Michael Curtis Mozer; Yoshua Bengio;
157	CoordX: Accelerating Implicit Neural Representation with A Split MLP Architecture Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we aim to accelerate inference and training of coordinate-based MLPs for implicit neural representations by proposing a new split MLP architecture, CoordX.	Ruofan Liang; Hongyi Sun; Nandita Vijaykumar;
158	COPA: Certifying Robust Policies for Offline Reinforcement Learning Against Poisoning Attacks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the first framework for certifiying robustness of offline reinforcement learning against poisoning attacks.	Fan Wu; Linyi Li; Huan Zhang; Bhavya Kailkhura; Krishnaram Kenthapadi; Ding Zhao; Bo Li;
159	COptiDICE: Offline Constrained Reinforcement Learning Via Stationary Distribution Correction Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present an offline constrained RL algorithm, which estimates the stationary distribution corrections of the optimal policy with respect to returns, while constraining the cost upper bound.	Jongmin Lee; Cosmin Paduraru; Daniel J Mankowitz; Nicolas Heess; Doina Precup; Kee-Eung Kim; Arthur Guez;
160	CosFormer: Rethinking Softmax In Attention Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A new linear transformer.	Qin Zhen; Weixuan Sun; Hui Deng; Dongxu Li; Yunshen Wei; Baohong Lv; Junjie Yan; Lingpeng Kong; Yiran Zhong;
161	CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Following this principle, we propose a new time series representation learning framework for time series forecasting named CoST, which applies contrastive learning methods to learn disentangled seasonal-trend representations.	Gerald Woo; Chenghao Liu; Doyen Sahoo; Akshat Kumar; Steven Hoi;
162	Counterfactual Plans Under Distributional Ambiguity Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This study the counterfactual plans under model uncertainty, in which the distribution of the model parameters is partially prescribed using only the first- and second-moment information.	Ngoc Bui; Duy Nguyen; Viet Anh Nguyen;
163	C-Planning: An Automatic Curriculum for Learning Goal-Reaching Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: An algorithm for goal-conditioned RL that uses an automatic curriculum of waypoints during exploration, derived from variational inference.	Tianjun Zhang; Benjamin Eysenbach; Ruslan Salakhutdinov; Sergey Levine; Joseph E. Gonzalez;
164	Creating Training Sets Via Weak Indirect Supervision Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we present a new weak supervision paradigm which automatically creates training sets for training a machine learning model given unlabeled dataset and indirect supervision sources.	Jieyu Zhang; Bohan Wang; Xiangchen Song; Yujing Wang; Yaming Yang; Jing Bai; Alexander Ratner;
165	Critical Points in Quantum Generative Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show using techniques from random matrix theory that, unlike typical neural networks, quantum generative models often have poor quality local minima.	Eric Ricardo Anschuetz;
166	CROP: Certifying Robust Policies for Reinforcement Learning Through Functional Smoothing Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose two particular types of robustness certification criteria: robustness of per-state actions and lower bound of cumulative rewards.	Fan Wu; Linyi Li; Zijian Huang; Yevgeniy Vorobeychik; Ding Zhao; Bo Li;
167	CrossBeam: Learning to Search in Bottom-Up Program Synthesis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose training a neural model to learn a hands-on search policy for bottom-up program synthesis, in an effort to tame the search space blowup.	Kensen Shi; Hanjun Dai; Kevin Ellis; Charles Sutton;
168	Cross-Domain Imitation Learning Via Optimal Transport Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study the use of Gromov-Wasserstein for cross-domain imitation learning We also release the set of considered benchmark tasks and implementations of our methods in order to further drive research in this challenging area.	Arnaud Fickinger; Samuel Cohen; Stuart Russell; Brandon Amos;
169	Cross-Domain Lossy Compression As Optimal Transport with An Entropy Bottleneck Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We consider the novel task of cross-distribution lossy compression and characterize it as an optimal transport problem under an entropy constraint, then provide experimental results to demonstrate the principles suggested by our theory.	Huan Liu; George Zhang; Jun Chen; Ashish J Khisti;
170	CrossFormer: A Versatile Vision Transformer Hinging on Cross-scale Attention Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To this end, we propose Cross-scale Embedding Layer (CEL) and Long Short Distance Attention (LSDA).	Wenxiao Wang; Lu Yao; Long Chen; Binbin Lin; Deng Cai; Xiaofei He; Wei Liu;
171	CrossMatch: Cross-Classifier Consistency Regularization for Open-Set Single Domain Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose a challenging and untouched problem: \textit{Open-Set Single Domain Generalization} (OS-SDG), where target domains include unseen categories out of source label space.	Ronghang Zhu; Sheng Li;
172	Crystal Diffusion Variational Autoencoder for Periodic Material Generation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A generative model for the 3D periodic structure of materials We also provide several standard datasets and evaluation metrics for the broader machine learning community.	Tian Xie; Xiang Fu; Octavian-Eugen Ganea; Regina Barzilay; Tommi S. Jaakkola;
173	Curriculum Learning As A Tool to Uncover Learning Principles in The Brain Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a novel approach to use curricula to identify principles by which a system learns.	Daniel R. Kepple; Rainer Engelken; Kanaka Rajan;
174	CURVATURE-GUIDED DYNAMIC SCALE NETWORKS FOR MULTI-VIEW STEREO Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper proposes a dynamic scale feature network to address the matching ambiguity problem in Multi-view stereo (MVS) and then designs an efficient MVS network to predict the depth maps.	Khang Truong Giang; Soohwan Song; Sungho Jo;
175	DAB-DETR: Dynamic Anchor Boxes Are Better Queries for DETR Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present in this paper a novel query formulation using dynamic anchor boxes for DETR and offer a deeper understanding of the role of queries in DETR.	Shilong Liu; Feng Li; Hao Zhang; Xiao Yang; Xianbiao Qi; Hang Su; Jun Zhu; Lei Zhang;
176	DARA: Dynamics-Aware Reward Augmentation in Offline Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we thus 1) formulate the offline dynamics adaptation by using (source) offline data collected from another dynamics to relax the requirement for the extensive (target) offline data, 2) characterize the dynamics shift problem in which prior offline methods does not scale well, and 3) derive a simple dynamics-aware reward augmentation (DARA) framework from both model-free and model-based offline settings.	Jinxin Liu; Zhang Hongyin; Donglin Wang;
177	Data-Driven Offline Optimization for Architecting Hardware Accelerators Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we develop such a data-driven offline optimization method for designing hardware accelerators, dubbed PRIME, that enjoys all of these properties.	Aviral Kumar; Amir Yazdanbakhsh; Milad Hashemi; Kevin Swersky; Sergey Levine;
178	Data Poisoning Won�t Save You From Facial Recognition Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Data poisoning and adversarial examples won’t protect users from facial recognition	Evani Radiya-Dixit; Sanghyun Hong; Nicholas Carlini; Florian Tramer;
179	D-CODE: Discovering Closed-form ODEs from Observed Trajectories Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we propose the Discovery of Closed-form ODE framework (D-CODE), which advances symbolic regression beyond the paradigm of supervised learning.	Zhaozhi Qian; Krzysztof Kacprzyk; Mihaela van der Schaar;
180	Dealing with Non-Stationarity in MARL Via Trust-Region Decomposition Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we introduce a novel notion, the – measurement, to explicitly measure the non-stationarity of a policy sequence, which can be further proved to be bounded by the KL-divergence of consecutive joint policies.	Wenhao Li; Xiangfeng Wang; Bo Jin; Junjie Sheng; Hongyuan Zha;
181	Decentralized Learning for Overparameterized Problems: A Multi-Agent Kernel Approximation Approach Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Decentralized optimization for overparameterized kernel learning	Prashant Khanduri; Haibo Yang; Mingyi Hong; Jia Liu; Hoi To Wai; Sijia Liu;
182	Declarative Nets That Are Equilibrium Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Choosing a kernelised generalised linear model as the inner problem of a DDN yields a DEQ with specific fixed weights.	Russell Tsuchida; Suk Yee Yong; Mohammad Ali Armin; Lars Petersson; Cheng Soon Ong;
183	Deconstructing The Inductive Biases of Hamiltonian Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we examine the inductive biases that make physics-inspired models successful in practice.	Nate Gruver; Marc Anton Finzi; Samuel Don Stanton; Andrew Gordon Wilson;
184	Decoupled Adaptation for Cross-Domain Object Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To deal with the challenges in cross-domain object detection, we propose D-adapt to decouple the adversarial adaptation and the training of the detector, and also decouple the category adaptation and the bounding box adaptation.	Junguang Jiang; Baixu Chen; Jianmin Wang; Mingsheng Long;
185	Deep Attentive Variational Inference Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Specifically, we propose deep attentive variational autoencoder and test it on a variety of established datasets.	Ifigeneia Apostolopoulou; Ian Char; Elan Rosenfeld; Artur Dubrawski;
186	Deep AutoAugment Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose Deep AutoAugment (DeepAA), a fully automated automated data augmentation methods that outperforms previous automated data augmentation methods.	Yu Zheng; Zhi Zhang; Shen Yan; Mi Zhang;
187	Deep Ensembling with No Overhead for Either Training or Testing: The All-Round Blessings of Dynamic Sparsity Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose an efficient ensemble learning framework FreeTickets via dynamic spasity, which is more efficient to train and inference than a single dense model, while matching the performance of the naive dense ensemble.	Shiwei Liu; Tianlong Chen; Zahra Atashgahi; Xiaohan Chen; Ghada Sokar; Elena Mocanu; Mykola Pechenizkiy; Zhangyang Wang; Decebal Constantin Mocanu;
188	Deep Learning Without Shortcuts: Shaping The Kernel with Tailored Rectifiers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we rectify this situation by developing a new type of transformation which is perfectly compatible with a variant of ReLUs — Leaky ReLUs.	Guodong Zhang; Aleksandar Botev; James Martens;
189	Deep Point Cloud Reconstruction Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose deep learning-based point cloud reconstruction algorithm	Jaesung Choe; ByeongIn Joung; Francois Rameau; Jaesik Park; In So Kweon;
190	Deep ReLU Networks Preserve Expected Length Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This article proves that, both on average and with high probability, randomly initialized ReLU networks with width larger than depth do not distort lengths and volumes.	Boris Hanin; Ryan S Jeong; David Rolnick;
191	Defending Against Image Corruptions Through Adversarial Augmentations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Our theoretically-supported method finds adversarial examples by optimizing over the weights of pre-trained autoencoders, and yields classifiers with improved robustness to image corruptions.	Dan Andrei Calian; Florian Stimberg; Olivia Wiles; Sylvestre-Alvise Rebuffi; Andr�s Gy�rgy; Timothy A Mann; Sven Gowal;
192	DEGREE: Decomposition Based Explanation for Graph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new decomposition based explanation for Graph Neural Networks.	Qizhang Feng; Ninghao Liu; Fan Yang; Ruixiang Tang; Mengnan Du; Xia Hu;
193	Delaunay Component Analysis for Evaluation of Data Representations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present Delaunay Component Analysis (DCA) framework for evaluation of learned data representations which anayzes geometric and topological properties of representation spaces using Delaunay graphs.	Petra Poklukar; Vladislav Polianskii; Anastasiia Varava; Florian T. Pokorny; Danica Kragic Jensfelt;
194	DemoDICE: Offline Imitation Learning with Supplementary Imperfect Demonstrations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Under this setting, we propose DemoDICE, which effectively utilizes imperfect demonstrations by matching the stationary distribution of a policy with experts’ distribution while penalizing its deviation from the overall demonstrations.	Geon-Hyeong Kim; Seokin Seo; Jongmin Lee; Wonseok Jeon; HyeongJoo Hwang; Hongseok Yang; Kee-Eung Kim;
195	Demystifying Batch Normalization in ReLU Networks: Equivalent Convex Optimization Models and Implicit Regularization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce an analytic framework based on convex duality to obtain exact and polynomial-time trainable convex representations of weight-decay regularized ReLU networks with BN.	Tolga Ergen; Arda Sahiner; Batu Ozturkler; John M. Pauly; Morteza Mardani; Mert Pilanci;
196	Demystifying Limited Adversarial Transferability in Automatic Speech Recognition Systems Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Uncover factors that limit transferability of the popular optimization attacks in the automatic speech recognition systems.	Hadi Abdullah; Aditya Karlekar; Vincent Bindschaedler; Patrick Traynor;
197	Denoising Likelihood Score Matching for Conditional Score-based Data Generation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we theoretically formulate a new training objective, called Denoising Likelihood Score Matching (DLSM) loss, for the classifier to match the gradients of the true log likelihood density.	Chen-Hao Chao; Wei-Fang Sun; Bo-Wun Cheng; Yi-Chen Lo; Chia-Che Chang; Yu-Lun Liu; Yu-Lin Chang; Chia-Ping Chen; Chun-Yi Lee;
198	DEPTS: Deep Expansion Learning for Periodic Time Series Forecasting Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we introduce a deep expansion learning framework, DEPTS, for PTS forecasting.	Wei Fan; Shun Zheng; Xiaohan Yi; Wei Cao; Yanjie Fu; Jiang Bian; Tie-Yan Liu;
199	DeSKO: Stability-Assured Robust Control with A Deep Stochastic Koopman Operator Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A robust learning control framework with guarantee stability based on deep stochastic Koopman operator models	Minghao Han; Jacob Euler-Rolle; Robert K. Katzschmann;
200	DictFormer: Tiny Transformer with Shared Dictionary Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose DictFormer with efficient shared dictionary to provide a compact, fast, and accurate transformer model.	Qian Lou; Ting Hua; Yen-Chang Hsu; Yilin Shen; Hongxia Jin;
201	Differentiable DAG Sampling Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new differentiable probabilistic model over DAGs (DP-DAG).	Bertrand Charpentier; Simon Kibler; Stephan G�nnemann;
202	Differentiable Expectation-Maximization for Set Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel set embedding function, a feed-forward network defined as the (differentiable) maximum-a-posterior estimate of the mixture, approximately attained by a few Expectation-Maximization steps.	Minyoung Kim;
203	Differentiable Prompt Makes Pre-trained Language Models Better Few-shot Learners Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A differentiable prompt learning method for few-shot NLP with optimized prompt templates as well as labels.	Ningyu Zhang; Luoqiu Li; Xiang Chen; Shumin Deng; Zhen Bi; Chuanqi Tan; Fei Huang; Huajun Chen;
204	Differentiable Scaffolding Tree for Molecule Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: make the molecular optimization problem differentiable at the structure level	Tianfan Fu; Wenhao Gao; Cao Xiao; Jacob Yasonik; Connor W. Coley; Jimeng Sun;
205	Differentially Private Fine-tuning of Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show that by combining recent advances in NLP, parameter-efficiency, privacy accounting, and using larger models, one can privately fine-tune models whose utility approaches that of non-private models.	Da Yu; Saurabh Naik; Arturs Backurs; Sivakanth Gopi; Huseyin A Inan; Gautam Kamath; Janardhan Kulkarni; Yin Tat Lee; Andre Manoel; Lukas Wutschitz; Sergey Yekhanin; Huishuai Zhang;
206	Differentially Private Fractional Frequency Moments Estimation with Polylogarithmic Space Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We prove that $\mathbb{F}_p$ sketch, a well-celebrated streaming algorithm for frequency moments estimation, is differentially private as is when $p\in(0, 1]$.	Lun Wang; Iosif Pinelis; Dawn Song;
207	Direct Then Diffuse: Incremental Unsupervised Skill Discovery for State Covering and Goal Reaching Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we build on the mutual information framework for skill discovery and introduce UPSIDE, which addresses the coverage-directedness trade-off in the following ways: 1) We design policies with a decoupled structure of a directed skill, trained to reach a specific region, followed by a diffusing part that induces a local coverage.	Pierre-Alexandre Kamienny; Jean Tarbouriech; Alessandro Lazaric; Ludovic Denoyer;
208	Discovering Latent Concepts Learned in BERT Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We approach interpretability from a model?s perspective by discovering and analyzing latent concepts learned in pre-trained models in an unsupervised fashion. We also release a novel BERT concept dataset (BCD) consisting of 198 concept labels and 1M annotated instances.	Fahim Dalvi; Abdul Rafae Khan; Firoj Alam; Nadir Durrani; Jia Xu; Hassan Sajjad;
209	Discovering Nonlinear PDEs from Scarce Data with Physics-encoded Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This work seeks to solve the data-driven governing equation discovery problem with a novel physics-encoded learning framework.	Chengping Rao; Pu Ren; Yang Liu; Hao Sun;
210	Discrete Representations Strengthen Vision Transformer Robustness Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a simple and effective architecture modification to ViT’s input layer with discrete token representations.	Chengzhi Mao; Lu Jiang; Mostafa Dehghani; Carl Vondrick; Rahul Sukthankar; Irfan Essa;
211	Disentanglement Analysis with Partial Information Decomposition Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We establish a framework to analyze information sharing in a multivariate representation with Partial Information Decomposition and propose a new disentanglement metric.	Seiya Tokui; Issei Sato;
212	DISSECT: Disentangled Simultaneous Explanations Via Concept Traversals Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel counterfactual explainability method that simultaneously satisfies several desirable qualities where other methods fail by training a generator, a discriminator, and a concept disentangler using the classifier?s signal.	Asma Ghandeharioun; Been Kim; Chun-Liang Li; Brendan Jou; Brian Eoff; Rosalind Picard;
213	Distilling GANs with Style-Mixed Triplets for X2I Translation with Limited Data Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: One transfer learning method generalizes varying kinds of conditional image synthesization tasks.	Yaxing Wang; Joost van de weijer; Lu Yu; SHANGLING JUI;
214	Distributional Decision Transformer for Hindsight Information Matching Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We generalize hindsight algorithms in RL, and propose Distributional Decision Transformer for information matching.	Hiroki Furuta; Yutaka Matsuo; Shixiang Shane Gu;
215	Distributionally Robust Fair Principal Components Via Geodesic Descents Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose a distributionally robust optimization problem for principal component analysis which internalizes a fairness criterion in the objective function.	Hieu Vu; Toan Tran; Man-Chung Yue; Viet Anh Nguyen;
216	Distributionally Robust Models with Parametric Likelihood Ratios Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We learn adversarial parametric reweightings of the training data to reliably train more robust models	Paul Michel; Tatsunori Hashimoto; Graham Neubig;
217	Distributional Reinforcement Learning with Monotonic Splines Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose to learn smooth continuous quantile functions represented by monotonic rational-quadratic splines, which also naturally solve the quantile crossing problem.	Yudong Luo; Guiliang Liu; Haonan Duan; Oliver Schulte; Pascal Poupart;
218	Distribution Compression in Near-Linear Time Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a simple algorithm for compressing an $n$-point summary of a probability distribution into a $\sqrt{n}$-point summary of comparable quality in $O(n \log^2 n)$ time.	Abhishek Shetty; Raaz Dwivedi; Lester Mackey;
219	Diurnal or Nocturnal? Federated Learning of Multi-branch Networks from Periodically Shifting Distributions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study a better modeling assumption for the periodical distribution shift in FL, and propose algorithms that learn better from the shifting distribution.	Chen Zhu; Zheng Xu; Mingqing Chen; Jakub Konecn�; Andrew Hard; Tom Goldstein;
220	DIVA: Dataset Derivative of A Learning Task Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Presents a method to optimize a dataset based on a notion of a dataset derivative that is computed in closed form using linearization	Yonatan Dukler; Alessandro Achille; Giovanni Paolini; Avinash Ravichandran; Marzia Polito; Stefano Soatto;
221	Dive Deeper Into Integral Pose Regression Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We do a deep dive on the inference and back-propagation of integral pose regression to better understand the causes behind the performance and training differences.	Kerui Gu; Linlin Yang; Angela Yao;
222	Diverse Client Selection for Federated Learning Via Submodular Maximization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The paper addresses a key challenge of selecting the most representative clients iteratively for federated learning through formulating it as a submodular optimization problem and developing efficient algorithms.	Ravikumar Balakrishnan; Tian Li; Tianyi Zhou; Nageen Himayat; Virginia Smith; Jeff Bilmes;
223	Divisive Feature Normalization Improves Image Recognition Performance in AlexNet Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: DIVISIVE FEATURE NORMALIZATION IMPROVES IMAGE RECOGNITION PERFORMANCE AND IN- CREASES MANIFOLD CAPACITY, SPARSITY, AND LOW-FREQUENCY REPRESENTATION IN DEEP NETS	Michelle Miller; SueYeon Chung; Kenneth D. Miller;
224	DKM: Differentiable K-Means Clustering Layer for Neural Network Compression Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel model compression scheme based on differentiable K-means layer, and it delivers the state-of-the-art results.	Minsik Cho; Keivan Alizadeh-Vahid; Saurabh Adya; Mohammad Rastegari;
225	Do Deep Networks Transfer Invariances Across Classes? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Study how well classifiers learn invariances in the imbalanced setting, and methods for improvement.	Allan Zhou; Fahim Tajwar; Alexander Robey; Tom Knowles; George J. Pappas; Hamed Hassani; Chelsea Finn;
226	Domain Adversarial Training: A Game Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A novel perspective on domain-adversarial training that leads to more stable and performant optimizers.	David Acuna; Marc T Law; Guojun Zhang; Sanja Fidler;
227	Do Not Escape From The Manifold: Discovering The Local Coordinates on The Latent Space of GANs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a method for finding local-geometry-aware traversal directions on the intermediate latent space of Generative Adversarial Networks (GANs).	Jaewoong Choi; Junho Lee; Changyeon Yoon; Jung Ho Park; Geonho Hwang; Myungjoo Kang;
228	Doubly Adaptive Scaled Algorithm for Machine Learning Using Second-Order Information Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Second-Order Method for Large Scale Machine Learning Tasks	Majid Jahani; Sergey Rusakov; Zheng Shi; Peter Richt�rik; Michael W. Mahoney; Martin Takac;
229	Do Users Benefit From Interpretable Vision? A User Study, Baseline, And Dataset Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Abstract: A variety of methods exist to explain image classification models. However, whether they provide any benefit to users over simply comparing various inputs and the model�s …	Leon Sixt; Martin Schuessler; Oana-Iuliana Popescu; Philipp Wei�; Tim Landgraf;
230	DR3: Value-Based Deep Reinforcement Learning Requires Explicit Regularization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that implicit regularization effects can lead to poor performance in value-based offline RL and propose an explicit regularizer to mitigate these effects.	Aviral Kumar; Rishabh Agarwal; Tengyu Ma; Aaron Courville; George Tucker; Sergey Levine;
231	DriPP: Driven Point Processes to Model Stimuli Induced Patterns in M/EEG Signals Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Model for patterns’ activation using temporal point processes to reveal stimulus-induced effects in brain electrophysiology.	C�dric Allain; Alexandre Gramfort; Thomas Moreau;
232	Dual Lottery Ticket Hypothesis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We articulate a Dual Lottery Ticket Hypothesis (DLTH) with a proposed training strategy Random Sparse Network to validate DLTH.	Yue Bai; Huan Wang; ZHIQIANG TAO; Kunpeng Li; Yun Fu;
233	Dynamics-Aware Comparison of Learned Reward Functions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a method for quantifying the similarity of learned reward functions without performing policy learning and evaluation.	Blake Wulfe; Logan Michael Ellis; Jean Mercat; Rowan Thomas McAllister; Adrien Gaidon;
234	Dynamic Token Normalization Improves Vision Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: The proposed DTN is a simple yet effective normalizer for vision transformers.	Wenqi Shao; Yixiao Ge; Zhaoyang Zhang; XUYUAN XU; Xiaogang Wang; Ying Shan; Ping Luo;
235	EE-Net: Exploitation-Exploration Neural Networks in Contextual Bandits Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we propose EE-Net, a neural-based bandit approach with a novel exploration strategy.	Yikun Ban; Yuchen Yan; Arindam Banerjee; Jingrui He;
236	Effective Model Sparsification By Scheduled Grow-and-Prune Methods Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we propose a novel scheduled grow-and-prune (GaP) methodology without having to pre-train a dense model.	Xiaolong Ma; Minghai Qin; Fei Sun; Zejiang Hou; Kun Yuan; Yi Xu; Yanzhi Wang; Yen-Kuang Chen; Rong Jin; Yuan Xie;
237	Effect of Scale on Catastrophic Forgetting in Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We find that large, pre-trained models are robust to catastrophic forgetting.	Vinay Venkatesh Ramasesh; Aitor Lewkowycz; Ethan Dyer;
238	Efficient Active Search for Combinatorial Optimization Problems Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose active search approaches for combinatorial optimization problems that search for solutions by adjusting a subset of (model) parameters to a single instance at test time.	Andr� Hottung; Yeong-Dae Kwon; Kevin Tierney;
239	Efficient and Differentiable Conformal Prediction with General Function Classes Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We generalize conformal prediction to learning multiple parameters within a general function class, to obtain an improved efficiency subject to valid coverage.	Yu Bai; Song Mei; Huan Wang; Yingbo Zhou; Caiming Xiong;
240	Efficient Computation of Deep Nonlinear Infinite-Width Neural Networks That Learn Features Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A new feature learning 8-width limit for deep nonlinear networks closes the performance gap between finite- and infinite-width neural networks previously left by NTK.	Greg Yang; Michael Santacroce; Edward J Hu;
241	Efficient Learning of Safe Driving Policy Via Human-AI Copilot Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we develop a novel human-in-the-loop learning method called Human-AI Copilot Optimization (HACO).	Quanyi Li; Zhenghao Peng; Bolei Zhou;
242	Efficient Neural Causal Discovery Without Acyclicity Constraints Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present ENCO, an efficient structure learning method that leverages observational and interventional data and scales to graphs with a thousand variables.	Phillip Lippe; Taco Cohen; Efstratios Gavves;
243	Efficient Self-supervised Vision Transformers for Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Achieving SoTA ImageNet linear probe task with 10 times higher throughput, using the synergy of a multi-stage Transformer architecture and a non-contrastive region-matching pre-training task.	Chunyuan Li; Jianwei Yang; Pengchuan Zhang; Mei Gao; Bin Xiao; Xiyang Dai; Lu Yuan; Jianfeng Gao;
244	Efficient Sharpness-aware Minimization for Improved Training of Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: An efficient sharpness aware minimizer that improves the generalization	Jiawei Du; Hanshu Yan; Jiashi Feng; Joey Tianyi Zhou; Liangli Zhen; Rick Siow Mong Goh; Vincent Tan;
245	Efficient Token Mixing for Transformers Via Adaptive Fourier Neural Operators Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Adaptive Fourier Neural Operators (AFNO) for scaling self-attention to high resolution images in vision transformers by establishing a link between operator learning and token mixing.	John Guibas; Morteza Mardani; Zongyi Li; Andrew Tao; Anima Anandkumar; Bryan Catanzaro;
246	Eigencurve: Optimal Learning Rate Schedule for SGD on Quadratic Objectives with Skewed Hessian Spectrums Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A learning rate schedule which achieves minimax optimal convergence rate (up to a constant) for SGD on quadratic objectives with skewed Hessian spectrums.	Rui Pan; Haishan Ye; Tong Zhang;
247	EigenGame Unloaded: When Playing Games Is Better Than Optimizing Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We improve the EigenGame algorithm by removing update bias, enabling further parallelism and better performance.	Ian Gemp; Brian McWilliams; Claire Vernade; Thore Graepel;
248	Embedded-model Flows: Combining The Inductive Biases of Model-free Deep Learning and Explicit Probabilistic Modeling Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce bijective transformations that embed domain-specific inductive biases in Normalizing Flow architectures.	Gianluigi Silvestri; Emily Fertig; Dave Moore; Luca Ambrogioni;
249	Emergent Communication at Scale Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work argues the importance of scaling up the emergent communication framework and investigates the impact of three scaling up aspects, namely the dataset, task complexity, and population size. We provide a first set of results for large populations solving complex tasks on realistic large-scale datasets, as well as an easy-to-use codebase to enable further experimentation.	Rahma Chaabouni; Florian Strub; Florent Altch�; Eugene Tarassov; Corentin Tallec; Elnaz Davoodi; Kory Wallace Mathewson; Olivier Tieleman; Angeliki Lazaridou; Bilal Piot;
250	Enabling Arbitrary Translation Objectives with Adaptive Tree Search Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: MCTS is used as a decoder for autoregressive and non-autoregressive machine translation models.	Wang Ling; Wojciech Stokowiec; Domenic Donato; Chris Dyer; Lei Yu; Laurent Sartran; Austin Matthews;
251	End-to-End Learning of Probabilistic Hierarchies on Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: End-to-end gradient-based hierarchical clustering on graphs by exploiting Markov chain theory leads to state-of-the-art results.	Daniel Z�gner; Bertrand Charpentier; Morgane Ayle; Sascha Geringer; Stephan G�nnemann;
252	Energy-Based Learning for Cooperative Games, with Applications to Valuation Problems in Machine Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: An energy-based treatment for cooperative games provides a decoupling perspective for Shapley value and others.	Yatao Bian; Yu Rong; Tingyang Xu; Jiaxiang Wu; Andreas Krause; Junzhou Huang;
253	Enhancing Cross-lingual Transfer By Manifold Mixup Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the cross-lingual manifold mixup method to improve the cross-lingual transfer.	Huiyun Yang; Huadong Chen; Hao Zhou; Lei Li;
254	EntQA: Entity Linking As Question Answering Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We frame entity linking as inverse open-domain question answering and solve the dilemma of having to predict mentions before entities.	Wenzheng Zhang; Wenyue Hua; Karl Stratos;
255	Entroformer: A Transformer-based Entropy Model for Learned Image Compression Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we propose a novel transformer-based entropy model, termed Entroformer, to capture long-range dependencies in probability distribution estimation effectively and efficiently.	Yichen Qian; Xiuyu Sun; Ming Lin; Zhiyu Tan; Rong Jin;
256	Environment Predictive Coding for Visual Navigation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce environment predicting coding, a self-supervised approach for learning environment-level representations for navigation-like tasks.	Santhosh Kumar Ramakrishnan; Tushar Nagarajan; Ziad Al-Halah; Kristen Grauman;
257	Equivariant and Stable Positional Encoding for More Powerful Graph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we revisit GNNs that allow using positional features of nodes given by positional encoding (PE) techniques such as Laplacian Eigenmap, Deepwalk, etc..	Haorui Wang; Haoteng Yin; Muhan Zhang; Pan Li;
258	Equivariant Self-Supervised Learning: Encouraging Equivariance in Representations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Imposing invariance to certain transformations (e.g. random resized cropping) and sensitivity to other transformations (e.g. four-fold rotations) learns better features.	Rumen Dangovski; Li Jing; Charlotte Loh; Seungwook Han; Akash Srivastava; Brian Cheung; Pulkit Agrawal; Marin Soljacic;
259	Equivariant Subgraph Aggregation Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a provably expressive graph learning framework based on representing graphs as multisets of subgraphs and processing them with an equivariant architecture.	Beatrice Bevilacqua; Fabrizio Frasca; Derek Lim; Balasubramaniam Srinivasan; Chen Cai; Gopinath Balamurugan; Michael M. Bronstein; Haggai Maron;
260	Equivariant Transformers for Neural Network Based Molecular Potentials Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel equivariant Transformer architecture for the prediction of molecular potentials and provide insights into the molecular representation through extensive analysis of the model’s attention weights.	Philipp Th�lke; Gianni De Fabritiis;
261	Escaping Limit Cycles: Global Convergence for Constrained Nonconvex-nonconcave Minimax Problems Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Under weak MVI we introduce a new extragradient-type algorithm that avoids limit cycles	Thomas Pethick; Puya Latafat; Panos Patrinos; Olivier Fercoq; Volkan Cevher;
262	Evading Adversarial Example Detection Defenses with Orthogonal Projected Gradient Descent Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We break four defenses that detect adversarial examples by introducing an improved attack technique that modifies the gradient before applying PGD.	Oliver Bryniarski; Nabeel Hingun; Pedro Pachuca; Vincent Wang; Nicholas Carlini;
263	Evaluating Distributional Distortion in Neural Language Modeling Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To address this gap, we develop a controlled evaluation scheme which uses generative models trained on natural data as artificial languages from which we can exactly compute sequence probabilities.	Benjamin LeBrun; Alessandro Sordoni; Timothy J. O’Donnell;
264	Evaluating Language-biased Image Classification Based on Semantic Compositionality Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We developed a benchmark test based on hierarchical semantic compositionality to evaluate the language-biased image classification of artificial models and evaluated the CLIP model using it.	Yoann Lemesle; Masataka Sawayama; Guillermo Valle-Perez; Maxime Adolphe; H�l�ne Sauz�on; Pierre-Yves Oudeyer;
265	Evaluating Model-Based Planning and Planner Amortization for Continuous Control Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We combine MPC with model-free RL and evaluate on continuous control tasks from scratch and in transfer settings; our results show that model-free RL is a strong baseline in single task settings and model-based methods shine in multi-goal tasks.	Arunkumar Byravan; Leonard Hasenclever; Piotr Trochim; Mehdi Mirza; Alessandro Davide Ialongo; Yuval Tassa; Jost Tobias Springenberg; Abbas Abdolmaleki; Nicolas Heess; Josh Merel; Martin Riedmiller;
266	Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We investigate the potential pitfalls of using MMD to evaluate graph generative models and propose recommendations for the practitioner on how to mitigate those challenges.	Leslie O’Bray; Max Horn; Bastian Rieck; Karsten Borgwardt;
267	Evidential Turing Processes Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: An original extension of evidential deep learning with neural processes and neural Turing machines makes it possible to attain both in-domain calibration and out-of-domain detection in a single model.	Melih Kandemir; Abdullah Akg�l; Manuel Haussmann; Gozde Unal;
268	EViT: Expediting Vision Transformers Via Token Reorganizations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose to reorganize attentive tokens in Vision Transformers to expedite inference speed.	Youwei Liang; Chongjian GE; Zhan Tong; Yibing Song; Jue Wang; Pengtao Xie;
269	Evolutionary Diversity Optimization with Clustering-based Selection for Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose EDO-CS, a new Evolutionary Diversity Optimization algorithm with Clustering-based Selection that can achieve a set of policies with both high quality and diversity efficiently.	Yutong Wang; Ke Xue; Chao Qian;
270	EXACT: Scalable Graph Neural Networks Training Via Extreme Activation Compression Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Based on the implementation, we propose a memory-efficient framework called “EXACT”, which for the first time demonstrate the potential and evaluate the feasibility of training GNNs with compressed activations.	Zirui Liu; Kaixiong Zhou; Fan Yang; Li Li; Rui Chen; Xia Hu;
271	Explainable GNN-Based Models Over Knowledge Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new family of graph neural network-based transformations of graph data that can be trained effectively and where all predictions can be explained symbolically as logical inferences in Datalog.	David Jaime Tena Cucala; Bernardo Cuenca Grau; Egor V. Kostylev; Boris Motik;
272	Explaining Point Processes By Learning Interpretable Temporal Logic Rules Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a principled method to learn a set of human-readable logic rules to explain temporal point processes.	Shuang Li; Mingquan Feng; Lu Wang; Abdelmajid Essofi; Yufeng Cao; Junchi Yan; Le Song;
273	Explanations of Black-Box Models Based on Directional Feature Interactions Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a bivariate explainer to explain directional feature interactions in black box models.	Aria Masoomi; Davin Hill; Zhonghui Xu; Craig P Hersh; Edwin K. Silverman; Peter J. Castaldi; Stratis Ioannidis; Jennifer Dy;
274	Exploring Extreme Parameter Compression for Pre-trained Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we aim to explore larger compression ratios for PLMs, among which tensor decomposition is a potential but under-investigated one.	Benyou Wang; Yuxin Ren; Lifeng Shang; Xin Jiang; Qun Liu;
275	Exploring Memorization in Adversarial Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper explores the memorization effect in adversarial training and analyzes its connections with model capacity, convergence, generalization, and especially robust overfitting of the adversarially trained models.	Yinpeng Dong; Ke Xu; Xiao Yang; Tianyu Pang; Zhijie Deng; Hang Su; Jun Zhu;
276	Exploring The Limits of Large Scale Pre-training Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We perform a systematic investigation of limits of large scale pre-training for few-shot and transfer learning in image recognition with a wide range of downstream tasks.	Samira Abnar; Mostafa Dehghani; Behnam Neyshabur; Hanie Sedghi;
277	Exposing The Implicit Energy Networks Behind Masked Language Models Via Metropolis–Hastings Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We interpret masked language models for sequences as energy based models and propose a tractable scheme inspired by Metropolis–Hasting Monte Carlo to draw samples from these models.	Kartik Goyal; Chris Dyer; Taylor Berg-Kirkpatrick;
278	ExT5: Towards Extreme Multi-Task Scaling for Transfer Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Using a suite of 107 NLP tasks, we show that massively multi-task pre-training can improve downstream performance on NLP tasks, overcoming trends of negative transfer between tasks while fine-tuning.	Vamsi Aribandi; Yi Tay; Tal Schuster; Jinfeng Rao; Huaixiu Steven Zheng; Sanket Vaibhav Mehta; Honglei Zhuang; Vinh Q. Tran; Dara Bahri; Jianmo Ni; Jai Gupta; Kai Hui; Sebastian Ruder; Donald Metzler;
279	FairCal: Fairness Calibration for Face Verification Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We calibrate face verification models for fairness, without use of the sensitive attribute, without the need for retraining. This leads to SOTA accuracy, fairness calibration, and equal FPRs across subgroups.	Tiago Salvador; Stephanie Cairns; Vikram Voleti; Noah Marshall; Adam M Oberman;
280	Fairness Guarantees Under Demographic Shift Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a strategy for designing classification algorithms that provide high-confidence fairness guarantees that remain valid if the distribution over observations changes after the trained model is deployed.	Stephen Giguere; Blossom Metevier; Yuriy Brun; Philip S. Thomas; Scott Niekum; Bruno Castro da Silva;
281	Fairness in Representation for Multilingual NLP: Insights from Controlled Experiments on Conditional Language Modeling Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We investigate performance disparity in multilingual NLP with Transformer conditional LMs, and find, in the context of computing, morphological complexity to be a byproduct of word segmentation and disparity arising therefrom unwarranted.	Ada Wan;
282	Fair Normalizing Flows Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new fair representation learning method based on normalizing flows which can bound the accuracy of any adversary trying to predict sensitive attributes.	Mislav Balunovic; Anian Ruoss; Martin Vechev;
283	FALCON: Fast Visual Concept Learning By Integrating Images, Linguistic Descriptions, and Conceptual Relations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a meta-learning framework for learning new visual concepts quickly, from just one or a few examples, guided by multiple naturally occurring data streams: simultaneously looking at images, reading sentences that describe the objects in the scene, and interpreting supplemental sentences that relate the novel concept with other concepts.	Lingjie Mei; Jiayuan Mao; Ziqi Wang; Chuang Gan; Joshua B. Tenenbaum;
284	Fast AdvProp Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we introduce Fast AdvProp, which aggressively revamps AdvProp’s costly training components, rendering the method nearly as cheap as the vanilla training setting.	Jieru Mei; Yucheng Han; Yutong Bai; Yixiao Zhang; Yingwei Li; Xianhang Li; Alan Yuille; Cihang Xie;
285	Fast Model Editing at Scale Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A computationally efficient approach for learning to edit the behavior of very large pre-trained language models (10 billion+ parameters)	Eric Mitchell; Charles Lin; Antoine Bosselut; Chelsea Finn; Christopher D Manning;
286	Fast Regression for Structured Inputs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that for a large class of structured inputs, such as combinations of low-rank matrices, sparse matrices, and Vandermonde matrices, regression can be approximately solved using runtime that is polynomial in p.	Raphael A Meyer; Cameron N Musco; Christopher P Musco; David Woodruff; Samson Zhou;
287	FastSHAP: Real-Time Shapley Value Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce FastSHAP, a new method for estimating Shapley values in a single forward pass using an explainer model that is learned via stochastic gradient optimization using a weighted least squares-like objective function.	Neil Jethani; Mukund Sudarshan; Ian Connick Covert; Su-In Lee; Rajesh Ranganath;
288	Fast Topological Clustering with Wasserstein Distance Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose a novel and computationally practical topological clustering method that clusters complex networks with intricate topology using principled theory from persistent homology and optimal transport.	Tananun Songdechakraiwut; Bryan M Krause; Matthew I Banks; Kirill V Nourski; Barry D Van Veen;
289	Feature Kernel Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A feature-learning perspective of (ensemble) Knowledge Distillation (KD) in Neural Networks to propose a new method (FKD), with both theoretical & experimental results demonstrating FKD’s advantages over standard KD baselines.	Bobby He; Mete Ozay;
290	FedBABU: Toward Enhanced Representation for Federated Image Classification Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel algorithm, FedBABU, which updates and aggregates only the body during federated training for enhanced representation.	Jaehoon Oh; SangMook Kim; Se-Young Yun;
291	FedPara: Low-rank Hadamard Product for Communication-Efficient Federated Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: New communication-efficient neural network parameterization for federated learning.	Nam Hyeon-Woo; Moon Ye-Bin; Tae-Hyun Oh;
292	Few-shot Learning Via Dirichlet Tessellation Ensemble Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We developed a novel geometric framework that greatly improves few-shot classification, based on Cluster-induced Voronoi Diagram (CIVD).	Chunwei Ma; Ziyun Huang; Mingchen Gao; Jinhui Xu;
293	FILIP: Fine-grained Interactive Language-Image Pre-Training Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce a large-scale Fine-grained Interacitve Language-Image Pretraining (FILIP) to achieve finer-level alignment through a new cross-modal late interaction mechanism, which can boost the performance on more grounded vision and language tasks. Furthermore, we construct a new large-scale image-text pair dataset called FILIP300M for pre-training.	Lewei Yao; Runhui Huang; Lu Hou; Guansong Lu; Minzhe Niu; Hang Xu; Xiaodan Liang; Zhenguo Li; Xin Jiang; Chunjing Xu;
294	Filling The G_ap_s: Multivariate Time Series Imputation By Graph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a graph neural network architecture for multivariate time series imputation and achieve state-of-the-art results on several benchmarks.	Andrea Cini; Ivan Marisca; Cesare Alippi;
295	FILM: Following Instructions in Language with Modular Methods Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a modular method for embodied instruction following; our method achieves SOTA on the ALFRED benchmark by a large margin while using less data by eschewing both expert trajectories and low-level instructions.	So Yeon Min; Devendra Singh Chaplot; Pradeep Kumar Ravikumar; Yonatan Bisk; Ruslan Salakhutdinov;
296	Finding An Unsupervised Image Segmenter in Each of Your Deep Generative Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose an entirely unsupervised method for foreground-background image segmentation based on automatically finding a direction in the latent space of a deep generative model.	Luke Melas-Kyriazi; Christian Rupprecht; Iro Laina; Andrea Vedaldi;
297	Finding Biological Plausibility for Adversarially Robust Features Via Metameric Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We suggest that the representations learned by an Adversarially Trained Network are aligned with Human Peripheral Computation	Anne Harrington; Arturo Deza;
298	Fine-grained Differentiable Physics: A Yarn-level Model for Fabrics Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Following this motivation, we propose a new differentiable fabrics model for composite materials such as cloths, where we dive into the granularity of yarns and model individual yarn physics and yarn-to-yarn interactions.	Deshan Gong; Zhanxing Zhu; Andrew J. Bulpitt; He Wang;
299	Finite-Time Convergence and Sample Complexity of Multi-Agent Actor-Critic Reinforcement Learning with Average Reward Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we establish the first finite-time convergence result of the actor-critic algorithm for fully decentralized multi-agent reinforcement learning (MARL) problems with average reward.	FNU Hairi; Jia Liu; Songtao Lu;
300	Fixed Neural Network Steganography: Train The Images, Not The Network Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A novel method for steganography based on adversarial perturbations.	Varsha Kishore; Xiangyu Chen; Yan Wang; Boyi Li; Kilian Q Weinberger;
301	FlexConv: Continuous Kernel Convolutions With Differentiable Kernel Sizes Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We provide a high bandwidth, alias-free convolutional kernel parameterization with learnable kernel size and constant parameter cost.	David W. Romero; Robert-Jan Bruintjes; Jakub Mikolaj Tomczak; Erik J Bekkers; Mark Hoogendoorn; Jan van Gemert;
302	Focus on The Common Good: Group Distributional Robustness Follows Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new and simple algorithm for the sub-population shift problem that enables learning of shared features and performed consistently well over several standard, and real-world, benchmarks of the problem.	Vihari Piratla; Praneeth Netrapalli; Sunita Sarawagi;
303	Fooling Explanations in Text Classifiers Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Our work shows that explanation methods in text classifiers are susceptible to imperceptible perturbations that alter the explanation outcomes without changing the predictions of the classifiers.	Adam Ivankay; Ivan Girardi; Chiara Marchiori; Pascal Frossard;
304	Fortuitous Forgetting in Connectionist Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce forget-and-relearn as a training paradigm where forgetting removes undesirable information and relearning bolsters useful features towards better generalization and compositionality.	Hattie Zhou; Ankit Vani; Hugo Larochelle; Aaron Courville;
305	FP-DETR: Detection Transformer Advanced By Fully Pre-training Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To mitigate the issue, we propose FP-DETR, a new method that Fully Pre-Trains an encoder-only transformer and smoothly fine-tunes it for object detection via a task adapter.	Wen Wang; Yang Cao; Jing Zhang; Dacheng Tao;
306	Frequency-aware SGD for Efficient Embedding Learning with Provable Benefits Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Specifically, we propose (Counter-based) Frequency-aware Stochastic Gradient Descent, which applies a frequency-dependent learning rate for each token, and exhibits provable speed-up compared to SGD when the token distribution is imbalanced.	Yan Li; Dhruv Choudhary; Xiaohan Wei; Baichuan Yuan; Bhargav Bhushanam; Tuo Zhao; Guanghui Lan;
307	From Intervention to Domain Transportation: A Novel Perspective to Optimize Recommendation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose and study a novel domain-transportation view for optimizing recommendation for information retrieval systems.	Da Xu; Yuting Ye; Chuanwei Ruan; Evren Korpeoglu; Sushant Kumar; Kannan Achan;
308	From Stars to Subgraphs: Uplifting Any GNN with Local Structure Awareness Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Our work stands between these two regimes: we introduce a general framework to uplift any MPNN to be more expressive, with limited scalability overhead and greatly improved practical performance.	Lingxiao Zhao; Wei Jin; Leman Akoglu; Neil Shah;
309	GATSBI: Generative Adversarial Training for Simulation-Based Inference Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Using generative adversarial networks for simulation-based inference	Poornima Ramesh; Jan-Matthis Lueckmann; Jan Boelts; �lvaro Tejero-Cantero; David S. Greenberg; Pedro J. Goncalves; Jakob H. Macke;
310	GDA-AM: ON THE EFFECTIVENESS OF SOLVING MIN-IMAX OPTIMIZATION VIA ANDERSON MIXING Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new minimax optimization framework,GDA-AM, that views the GDA dynamics as a fixed-point iteration and solves it using Anderson Mixing to converge to the local minimax.	Huan He; Shifan Zhao; Yuanzhe Xi; Joyce Ho; Yousef Saad;
311	GeneDisco: A Benchmark for Experimental Design in Drug Discovery Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A comprehensive benchmark for batch active learning in drug discovery.	Arash Mehrjou; Ashkan Soleymani; Andrew Jesson; Pascal Notin; Yarin Gal; Stefan Bauer; Patrick Schwab;
312	Generalization of Neural Combinatorial Solvers Through The Lens of Adversarial Robustness Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study the generalization of combinatorial optimization w.r.t. to adversarial attacks since current evaluation protocols are too optimistic and we show that neural solvers are indeed vulnerable under label-preserving perturbations.	Simon Geisler; Johanna Sommer; Jan Schuchardt; Aleksandar Bojchevski; Stephan G�nnemann;
313	Generalization of Overparametrized Deep Neural Network Under Noisy Observations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Study the generalization of overparametrized deep neural network with relu activation function with noisy dataset.	Namjoon Suh; Hyunouk Ko; Xiaoming Huo;
314	Generalization Through The Lens of Leave-One-Out Error Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we instead revisit the concept of leave-one-out error to measure the generalization ability of deep and wide models.	Gregor Bachmann; Thomas Hofmann; Aurelien Lucchi;
315	Generalized Demographic Parity for Group Fairness Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This work aims to generalize demographic parity to continuous sensitive attributes while preserving tractable computation.	Zhimeng Jiang; Xiaotian Han; Chao Fan; Fan Yang; Ali Mostafavi; Xia Hu;
316	Generalized Kernel Thinning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: By generalizing kernel thinning, we develop new tools for compressing distributions more effectively than i.i.d. sampling and establish new dimension-free, $O(\sqrt{\log n/n})$ improvements over the usual $n^{-1/4}$ Monte Carlo integration error.	Raaz Dwivedi; Lester Mackey;
317	Generalized Natural Gradient Flows in Hidden Convex-Concave Games and GANs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper studies the generalized Gradient Descent-Ascent (GDA) flow in a large class of non-convex non-concave Zero-Sum games dubbed Hidden Convex-Concave games, a class of games that includes GANs.	Andjela Mladenovic; Iosif Sakos; Gauthier Gidel; Georgios Piliouras;
318	Generalized Rectifier Wavelet Covariance Models for Texture Synthesis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper presents a model for texture synthesis, built on a wavelet-based representation of images.	Antoine Brochard; Sixin Zhang;
319	Generalizing Few-Shot NAS with Gradient Matching Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we propose a gradient matching score (GM) that leverages gradient information at the shared weight for making informed splitting decisions.	Shoukang Hu; Ruochen Wang; Lanqing HONG; Zhenguo Li; Cho-Jui Hsieh; Jiashi Feng;
320	Generating Videos with Dynamics-aware Implicit Generative Adversarial Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We make video generation scalable leveraging implicit neural representations.	Sihyun Yu; Jihoon Tack; Sangwoo Mo; Hyunsu Kim; Junho Kim; Jung-Woo Ha; Jinwoo Shin;
321	Generative Modeling with Optimal Transport Maps Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: While optimal transport cost serves as the loss for popular generative models, we demonstrate that the optimal transport map can be used as the generative model itself.	Litu Rout; Alexander Korotin; Evgeny Burnaev;
322	Generative Planning for Temporally Coordinated Exploration in Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Temporally coordinated exploration in reinforcement learning using Generative Planning Method.	Haichao Zhang; Wei Xu; Haonan Yu;
323	Generative Principal Component Analysis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study the problem of principal component analysis with generative modeling assumptions, and provide a corresponding efficient algorithm with provable guarantees.	Zhaoqiang Liu; Jiulong Liu; Subhroshekhar Ghosh; Jun Han; Jonathan Scarlett;
324	Generative Pseudo-Inverse Memory Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Generative Pseudo-Inverse Memory (GPM), a class of deep generative memory models that are fast to write in and read out.	Kha Pham; Hung Le; Man Ngo; Truyen Tran; Bao Ho; Svetha Venkatesh;
325	Geometric and Physical Quantities Improve E(3) Equivariant Message Passing Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We generalise equivariant graph networks such that node and edge updates are able to leverage covariant information.	Johannes Brandstetter; Rob Hesselink; Elise van der Pol; Erik J Bekkers; Max Welling;
326	Geometric Transformers for Protein Interface Contact Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a geometry-evolving graph transformer for 3D protein structures and employ it to achieve state-of-the-art precision for predicting inter-protein residue-residue contacts in challenging protein complex targets.	Alex Morehead; Chen Chen; Jianlin Cheng;
327	GiraffeDet: A Heavy-Neck Paradigm for Object Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: we propose a novel heavy-neck paradigm(GiraffeDet) for detection task, which allows detectors to process the high-level categorical information and low-level spatial information uniformly, making it more effective in detection tasks.	yiqi jiang; Zhiyu Tan; Junyan Wang; Xiuyu Sun; Ming Lin; Hao Li;
328	Givens Coordinate Descent Methods for Rotation Matrix Learning in Trainable Embedding Indexes Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learning orthonormal matrix in neural networks via Givens rotations	Yunjiang Jiang; Han Zhang; Yiming Qiu; Yun Xiao; Bo Long; Wen-Yun Yang;
329	GLASS: GNN with Labeling Tricks for Subgraph Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: With this insight, we introduce an expressive and scalable labeling trick, namely max-zero-one, and propose GLASS (GNN with LAbel for SubgraphS).	Xiyuan Wang; Muhan Zhang;
330	Global Convergence of Multi-Agent Policy Gradient in Markov Potential Games Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Convergence of policy gradient in a class of MDPs called Markov Potential Games in which cooperation is desired.	Stefanos Leonardos; Will Overman; Ioannis Panageas; Georgios Piliouras;
331	GNN Is A Counter? Revisiting GNN for Question Answering Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Counting is essential for reasoning and our simplistic graph neural counter is efficient and effective for QA tasks.	Kuan Wang; Yuyu Zhang; Diyi Yang; Le Song; Tao Qin;
332	GNN-LM: Language Modeling Based on Global Contexts Via GNN Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Inspired by the notion that “{\it to copy is easier than to memorize}“, in this work, we introduce GNN-LM, which extends vanilla neural language model (LM) by allowing to reference similar contexts in the entire training corpus.	Yuxian Meng; Shi Zong; Xiaoya Li; Xiaofei Sun; Tianwei Zhang; Fei Wu; Jiwei Li;
333	Goal-Directed Planning Via Hindsight Experience Replay Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper presents an extension of AlphaZero to tackle sparse reward goal-based tasks	Lorenzo Moro; Amarildo Likmeta; Enrico Prati; Marcello Restelli;
334	Gradient Importance Learning for Incomplete Observations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we introduce the gradient importance learning (GIL) method to train multilayer perceptrons (MLPs) and long short-term memories (LSTMs) to directly perform inference from inputs containing missing values without imputation.	Qitong Gao; Dong Wang; Joshua David Amason; Siyang Yuan; Chenyang Tao; Ricardo Henao; Majda Hadziahmetovic; Lawrence Carin; Miroslav Pajic;
335	Gradient Information Matters in Policy Optimization By Back-propagating Through Model Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Considering the gradient information in the model learning is crucial for the model-based policy optimization according to our theoritical results. Motivated by such conclusion, we design a novel DDPPO algorithm that can achieve the SOTA performance.	Chongchong Li; Yue Wang; Wei Chen; Yuting Liu; Zhi-Ming Ma; Tie-Yan Liu;
336	Gradient Matching for Domain Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose to learn features that are invariant across domains by maximizing the gradient inner product between domains.	Yuge Shi; Jeffrey Seely; Philip Torr; Siddharth N; Awni Hannun; Nicolas Usunier; Gabriel Synnaeve;
337	Gradient Step Denoiser for Convergent Plug-and-Play Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a performant Plug-and-Play image restoration algorithm that theoretically converges with an exact gradient step deep denoiser.	Samuel Hurault; Arthur Leclaire; Nicolas Papadakis;
338	GradMax: Growing Neural Networks Using Gradient Information Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a method that adds new neurons during training without impacting what is already learned, while improving the training dynamics.	Utku Evci; Bart van Merrienboer; Thomas Unterthiner; Fabian Pedregosa; Max Vladymyrov;
339	GradSign: Model Performance Inference with Theoretical Insights Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Model performance inference inspired by sample-wise optimization landscape analysis	Zhihao Zhang; Zhihao Jia;
340	GRAND++: Graph Neural Diffusion with A Source Term Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose GRAND++ for deep graph learning with limited labeled training data	Matthew Thorpe; Tan Minh Nguyen; Hedi Xia; Thomas Strohmer; Andrea Bertozzi; Stanley Osher; Bao Wang;
341	Granger Causal Inference on DAGs Identifies Genomic Loci Regulating Transcription Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show how to extend Granger causality to DAG-structured dynamical systems using graph neural networks, applying it to infer noncoding regions involved in gene regulation.	Alexander P Wu; Rohit Singh; Bonnie Berger;
342	Graph-Augmented Normalizing Flows for Anomaly Detection of Multiple Time Series Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We hypothesize that anomalies occur in low density regions of a distribution and explore the use of normalizing flows for unsupervised anomaly detection, because of their superior quality in density estimation.	Enyan Dai; Jie Chen;
343	Graph Auto-Encoder Via Neighborhood Wasserstein Reconstruction Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study unsupervised graph representation learning and propose a novel decoder based on neighborhood reconstruction with Wasserstein distance to facilitate the GNN encoding of entire neighborhood information beyond direct links.	Mingyue Tang; Pan Li; Carl Yang;
344	Graph-Enhanced Exploration for Goal-oriented Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose G2RL, a new goal-oriented RL that leverages the state-transition graph for effective exploration and efficient training.	Jiarui Jin; Sijin Zhou; Weinan Zhang; Tong He; Yong Yu; Rasool Fakoor;
345	GraphENS: Neighbor-Aware Ego Network Synthesis for Class-Imbalanced Node Classification Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we hypothesize that overfitting to the neighbor sets of minor class due to message passing is a major challenge for class-imbalanced node classification.	Joonhyung Park; Jaeyun Song; Eunho Yang;
346	Graph-Guided Network for Irregularly Sampled Multivariate Time Series Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Here, we introduce RAINDROP, a graph-guided network for learning representations of irregularly sampled multivariate time series.	Xiang Zhang; Marko Zeman; Theodoros Tsiligkaridis; Marinka Zitnik;
347	Graph-less Neural Networks: Teaching Old MLPs New Tricks Via Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Distill knowledge from GNNs to MLPs to accelerate model inference and facilitate deployment for large-scale applications	Shichang Zhang; Yozen Liu; Yizhou Sun; Neil Shah;
348	Graph Neural Networks with Learnable Structural and Positional Representations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel GNN architecture (LSPE) which decouples structural and positional representations to make easy for the network to learn the two essential properties.	Vijay Prakash Dwivedi; Anh Tuan Luu; Thomas Laurent; Yoshua Bengio; Xavier Bresson;
349	Graphon Based Clustering and Testing of Networks: Algorithms and Theory Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we propose methods for clustering multiple graphs, without vertex correspondence, that are inspired by the recent literature on estimating graphons—symmetric functions corresponding to infinite vertex limit of graphs.	Mahalakshmi Sabanayagam; Leena Chennuru Vankadara; Debarghya Ghoshdastidar;
350	Graph-Relational Domain Adaptation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we relax such uniform alignment by using a domain graph to encode domain adjacency, e.g., a graph of states in the US with each state as a domain and each edge indicating adjacency, thereby allowing domains to align flexibly based on the graph structure.	Zihao Xu; Hao He; Guang-He Lee; Bernie Wang; Hao Wang;
351	GreaseLM: Graph REASoning Enhanced Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose GreaseLM, a new model that fuses encoded representations from pretrained LMs and GNNs over multiple layers of modality interaction operations, allowing both modalities to bidirectionally inform the representation of the other.	Xikun Zhang; Antoine Bosselut; Michihiro Yasunaga; Hongyu Ren; Percy Liang; Christopher D Manning; Jure Leskovec;
352	Group-based Interleaved Pipeline Parallelism for Large-scale DNN Training Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we propose a novel pipeline named WPipe, which achieves better memory efficiency and fresher weight updates.	PengCheng Yang; Xiaoming Zhang; Wenpeng Zhang; Ming Yang; Hong Wei;
353	Group Equivariant Neural Posterior Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We describe a method to incorporate group equivariances into neural posterior estimation.	Maximilian Dax; Stephen R Green; Jonathan Gair; Michael Deistler; Bernhard Sch�lkopf; Jakob H. Macke;
354	Half-Inverse Gradients for Physical Deep Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: By proposing a novel Jacobian-based optimizer, we question the current practice of using the state-of-the-art gradient-based methods for the optimization of neural networks with physics objectives.	Patrick Schnell; Philipp Holl; Nils Thuerey;
355	Heteroscedastic Temporal Variational Autoencoder For Irregularly Sampled Time Series Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a new deep learning architecture for probabilistic interpolation of irregularly sampled time series.	Satya Narayan Shukla; Benjamin Marlin;
356	Hidden Convexity of Wasserstein GANs: Interpretable Generative Models with Closed-Form Solutions Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We demonstrate that Wasserstein GANs with two-layer discriminators and a variety of generators are equivalent to convex optimization problems or convex-concave games, allowing for global optimization in polynomial time and improved interpretability.	Arda Sahiner; Tolga Ergen; Batu Ozturkler; Burak Bartan; John M. Pauly; Morteza Mardani; Mert Pilanci;
357	Hidden Parameter Recurrent State Space Models For Changing Dynamics Scenarios Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A new formalism for extending Recurrent State Space Models (RSSMs) to changing dynamics scenarios.	Vaisakh Shaj; Dieter B�chler; Rohit Sonker; Philipp Becker; Gerhard Neumann;
358	Hierarchical Few-Shot Imitation with Skill Transition Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a new algorithm (FIST) which extracts skills from offline data and adapts them in few-shot to solve unseen complex long-horizon tasks by utilizing an inverse skill dynamics model and semi-parametric imitation.	Kourosh Hakhamaneshi; Ruihan Zhao; Albert Zhan; Pieter Abbeel; Michael Laskin;
359	Hierarchical Variational Memory for Few-shot Learning Across Domains Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a hierarchical memory that stores features at different semantic levels for the cross-domain few-shot learning.	Yingjun Du; Xiantong Zhen; Ling Shao; Cees G. M. Snoek;
360	High Probability Bounds for A Class of Nonconvex Algorithms with AdaGrad Stepsize Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose a new, simplified high probability analysis of AdaGrad for smooth, non-convex problems.	Ali Kavis; Kfir Yehuda Levy; Volkan Cevher;
361	High Probability Generalization Bounds for Minimax Problems with Fast Rates Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we provide improved generalization analyses for almost all existing generalization measures of minimax problems, which enables the minimax problems to establish sharper bounds of order $\mathcal{O}\left( 1/n \right)$, significantly, with high probability.	Shaojie Li; Yong Liu;
362	Hindsight Foresight Relabeling for Meta-Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present HFR, a relabeling method that can be applied to meta-reinforcement learning to boost sample efficiency and performance.	Michael Wan; Jian Peng; Tanmay Gangwani;
363	Hindsight Is 20/20: Leveraging Past Traversals to Aid 3D Perception Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To this end, we propose a novel end-to-end trainable Hindsight framework to extract this contextual information from past traversals and store it in an easy-to-query data structure, which can then be leveraged to aid future 3D object detection of the same scene.	Yurong You; Katie Z Luo; Xiangyu Chen; Junan Chen; Wei-Lun Chao; Wen Sun; Bharath Hariharan; Mark Campbell; Kilian Q Weinberger;
364	Hindsight: Posterior-guided Training of Retrievers for Improved Open-ended Generation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We use a posterior-guide retriever to train a retrieval-augmented generation that performs well on open-ended one-to-many generation tasks.	Ashwin Paranjape; Omar Khattab; Christopher Potts; Matei Zaharia; Christopher D Manning;
365	Hot-Refresh Model Upgrades with Regression-Free Compatible Training in Image Retrieval Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We for the first time study the model regression problem in hot-refresh model upgrades of image retrieval systems with compatible representation learning.	Binjie Zhang; Yixiao Ge; Yantao Shen; Yu Li; Chun Yuan; XUYUAN XU; Yexin Wang; Ying Shan;
366	How Attentive Are Graph Attention Networks? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We identify that Graph Attention Networks (GAT) compute a very weak form of attention. We show its empirical implications and propose a fix.	Shaked Brody; Uri Alon; Eran Yahav;
367	How Did The Model Change? Efficiently Assessing Machine Learning API Shifts Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We systematically study real world ML APIs’ performance shifts due to API updates/retraining and propose a framework to efficiently estimate those shifts.	Lingjiao Chen; Matei Zaharia; James Zou;
368	How Does SimSiam Avoid Collapse Without Negative Samples? Towards A Unified Understanding of Progress in SSL Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: After refuting their claims, we introduce vector decomposition for analyzing the collapse based on the gradient analysis of $l_2$ normalized vector.	Chaoning Zhang; Kang Zhang; Chenshuang Zhang; Trung X. Pham; Chang D. Yoo; In So Kweon;
369	How Do Vision Transformers Work? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show that (1) multi-head self-attentions (MSAs) for computer vision flatten the loss landscapes, (2) MSAs are low-pass filters as opposed to Convs, and (3) MSAs at the end of a stage significantly improve the accuracy.	Namuk Park; Songkuk Kim;
370	How Low Can We Go: Trading Memory for Error in Low-Precision Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Given a dataset and a memory budget, we use matrix factorization and active learning to efficiently pick the perfect low-precision configuration for a neural network.	Chengrun Yang; Ziyang Wu; Jerry Chee; Christopher De Sa; Madeleine Udell;
371	How Many Degrees of Freedom Do We Need to Train Deep Networks: A Loss Landscape Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A variety of recent works, spanning pruning, lottery tickets, and training within random subspaces, have shown that deep neural networks can be trained using far fewer degrees of freedom than the total number of parameters. We analyze this phenomenon for random subspaces by first examining the success probability of hitting a training loss sublevel set when training within a random subspace of a given training dimensionality.	Brett W Larsen; Stanislav Fort; Nic Becker; Surya Ganguli;
372	How Much Can CLIP Benefit Vision-and-Language Tasks? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To further study the advantage brought by CLIP, we propose to use CLIP as the visual encoder in various V&L models in two typical scenarios: 1) plugging CLIP into task-specific fine-tuning; 2) combining CLIP with V&L pre-training and transferring to downstream tasks.	Sheng Shen; Liunian Harold Li; Hao Tan; Mohit Bansal; Anna Rohrbach; Kai-Wei Chang; Zhewei Yao; Kurt Keutzer;
373	How to Deal with Missing Data in Supervised Deep Learning? Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Marginalize over missing values in supervised learning using deep latent variable models.	Niels Bruun Ipsen; Pierre-Alexandre Mattei; Jes Frellsen;
374	How to Inject Backdoors with Better Consistency: Logit Anchoring on Clean Data Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel logit anchoring approach for better global and instance-wise consistency in backdoor learning.	Zhiyuan Zhang; Lingjuan Lyu; Weiqiang Wang; Lichao Sun; Xu Sun;
375	How to Robustify Black-Box ML Models? A Zeroth-Order Optimization Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a general notion of defensive operation that can be applied to black-box models, and design it through the lens of denoised smoothing (DS), a first-order (FO) certified defense technique.	Yimeng Zhang; Yuguang Yao; Jinghan Jia; Jinfeng Yi; Mingyi Hong; Shiyu Chang; Sijia Liu;
376	How to Train Your MAML to Excel in Few-Shot Classification Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we point out several key facets of how to train MAML to excel in few-shot classification.	Han-Jia Ye; Wei-Lun Chao;
377	HTLM: Hyper-Text Pre-Training and Prompting of Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We unlock new state-of-the-art ways of priming and automatically generating prompts by pre-training on simplified HTML.	Armen Aghajanyan; Dmytro Okhonko; Mike Lewis; Mandar Joshi; Hu Xu; Gargi Ghosh; Luke Zettlemoyer;
378	HyAR: Addressing Discrete-Continuous Action Reinforcement Learning Via Hybrid Action Representation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we propose Hybrid Action Representation (HyAR) to learn a compact and decodable latent representation space for the original hybrid action space.	Boyan Li; Hongyao Tang; YAN ZHENG; Jianye HAO; Pengyi Li; Zhen Wang; Zhaopeng Meng; LI Wang;
379	Hybrid Local SGD for Federated Learning with Heterogeneous Communications Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In light of both device-to-device (D2D) and device-to-server (D2E) cooperation opportunities in modern communication networks, this paper proposes a new federated optimization algorithm dubbed hybrid local SGD (HL-SGD) in FL settings where devices are grouped into a set of disjoint clusters with high D2D communication bandwidth.	Yuanxiong Guo; Ying Sun; Rui Hu; Yanmin Gong;
380	Hybrid Memoised Wake-Sleep: Approximate Inference at The Discrete-Continuous Interface Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Here, we propose Hybrid Memoised Wake-Sleep (HMWS), an algorithm for effective inference in such hybrid discrete-continuous models.	Tuan Anh Le; Katherine M. Collins; Luke Hewitt; Kevin Ellis; Siddharth N; Samuel Gershman; Joshua B. Tenenbaum;
381	Hybrid Random Features Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new class of random feature methods for softmax and Gaussian kernel estimation that are adaptable to provide particularly accurate approximation in the desired regions of interest.	Krzysztof Marcin Choromanski; Han Lin; Haoxian Chen; Arijit Sehanobish; Yuanzhe Ma; Deepali Jain; Jake Varley; Andy Zeng; Michael S Ryoo; Valerii Likhosherstov; Dmitry Kalashnikov; Vikas Sindhwani; Adrian Weller;
382	HyperDQN: A Randomized Exploration Method for Deep Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We design a practical randomized exploration method to address the sample efficiency issue in online reinforcement learning.	Ziniu Li; Yingru Li; Yushun Zhang; Tong Zhang; Zhi-Quan Luo;
383	IFlood: A Stable and Effective Regularizer Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel regularizer named iFlood, which encourages the trained models to better fit the under-fitted instances while suppressing the confidence on over-fitted ones.	Yuexiang Xie; Zhen WANG; Yaliang Li; Ce Zhang; Jingren Zhou; Bolin Ding;
384	IFR-Explore: Learning Inter-object Functional Relationships in 3D Indoor Scenes Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We formulate a novel problem of learning inter-object functional relationships in 3D indoor environments and propose a novel method that combines prior knowledge modeling and interactive policy learning to solve the task. We create a new dataset based on the AI2Thor and PartNet datasets and perform extensive experiments that prove the effectiveness of our proposed method.	QI LI; Kaichun Mo; Yanchao Yang; Hang Zhao; Leonidas Guibas;
385	Igeood: An Information Geometry Approach to Out-of-Distribution Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a flexible and effective out-of-distribution detection method by building on the Fisher-Rao distance between probability distributions.	Eduardo Dadalto Camara Gomes; Florence Alberge; Pierre Duhamel; Pablo Piantanida;
386	IGLU: Efficient GCN Training Via Lazy Updates Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: IGLU is a novel lazy update-based optimization technique for accelerated GCN training with provable convergence guarantees	S Deepak Narayanan; Aditya Sinha; Prateek Jain; Purushottam Kar; SUNDARARAJAN SELLAMANICKAM;
387	Illiterate DALL$\cdot$E Learns to Compose Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To learn compositional slot-based representation of an image and perform slot composition for zero-shot novel image generation.	Gautam Singh; Fei Deng; Sungjin Ahn;
388	Image BERT Pre-training with Online Tokenizer Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a self-supervised framework iBOT that can perform masked image modeling with an online tokenizer, achieving the state-of-the-art results in downstream tasks.	Jinghao Zhou; Chen Wei; Huiyu Wang; Wei Shen; Cihang Xie; Alan Yuille; Tao Kong;
389	Imbedding Deep Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Invariant imbedding solution for (Bolza) optimal control problem derived and proved to yield new architectures of imbedded deep neural networks.	Andrew Corbett; Dmitry Kangin;
390	Imitation Learning By Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: For deterministic experts, you can do imitation learning by calling an RL solver once, with a stationary reward signal.	Kamil Ciosek;
391	Implicit Bias of Adversarial Training for Deep Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide theoretical understandings of the implicit bias imposed by adversarial training for homogeneous deep neural networks without explicit regularization.	Bochen Lv; Zhanxing Zhu;
392	Implicit Bias of MSE Gradient Optimization in Underparameterized Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Underparameterized networks optimizing MSE learn eigenfunctions of an NTK integral operator at rates corresponding to their eigenvalues.	Benjamin Bowman; Guido Montufar;
393	Implicit Bias of Projected Subgradient Method Gives Provable Robust Recovery of Subspaces of Unknown Codimension Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study the robust subspace recovery problem when subspace codimension is unknown.	Paris Giampouras; Benjamin David Haeffele; Rene Vidal;
394	Improved Deterministic L2 Robustness on CIFAR-10 and CIFAR-100 Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Improving provable robustness of 1 Lipschitz CNNs by relaxing orthogonalization of last layer, certificate regularization and a novel activation function.	Sahil Singla; Surbhi Singla; Soheil Feizi;
395	Improving Federated Learning Face Recognition Via Privacy-Agnostic Clusters Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To resolve the privacy-utility paradox, this work proposes PrivacyFace, a framework largely improves the federated learning face recognition via communicating auxiliary and privacy-agnostic information among clients.	Qiang Meng; Feng Zhou; Hainan Ren; Tianshu Feng; Guochao Liu; Yuanqing Lin;
396	Improving Mutual Information Estimation with Annealed and Energy-Based Bounds Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We derive new annealed importance sampling and energy-based bounds, resulting in vastly more accurate estimates of mutual information.	Rob Brekelmans; Sicong Huang; Marzyeh Ghassemi; Greg Ver Steeg; Roger Baker Grosse; Alireza Makhzani;
397	Improving Non-Autoregressive Translation Models Without Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Improving the CMLM non-autoregressive machine translation model so it trains without knowledge distillation and achieves SOTA BLEU score on both raw and distilled dataset	Xiao Shi Huang; Felipe Perez; Maksims Volkovs;
398	In A Nutshell, The Human Asked for This: Latent Goals for Following Temporal Specifications Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Inducing architectures to generate low-dimensional representations of their current goal processing observations and instructions together yields stronger out-of-distribution generalisation	Borja G. Le�n; Murray Shanahan; Francesco Belardinelli;
399	Increasing The Cost of Model Extraction with Calibrated Proof of Work Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose to make model extraction more difficult by requiring users to complete a callibrated proof-of-work before they can read predictions from a machine learning model exposed via a public API.	Adam Dziedzic; Muhammad Ahmad Kaleem; Yu Shen Lu; Nicolas Papernot;
400	Incremental False Negative Detection for Contrastive Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper explores the effect of false negative samples in self-supervised contrastive learning and introduce a framework to incrementally detect and explicitly remove the false negatives.	Tsai-Shien Chen; Wei-Chih Hung; Hung-Yu Tseng; Shao-Yi Chien; Ming-Hsuan Yang;
401	Independent SE(3)-Equivariant Models for End-to-End Rigid Protein Docking Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We perform rigid protein docking using a novel independent SE(3)-equivariant message passing mechanism that guarantees the same resulting protein complex independent of the initial placement of the two 3D structures.	Octavian-Eugen Ganea; Xinyuan Huang; Charlotte Bunne; Yatao Bian; Regina Barzilay; Tommi S. Jaakkola; Andreas Krause;
402	InfinityGAN: Towards Infinite-Pixel Image Synthesis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: InfinityGAN learns to synthesize arbitrary-sized images with limited resources and enables multiple new applications.	Chieh Hubert Lin; Hsin-Ying Lee; Yen-Chi Cheng; Sergey Tulyakov; Ming-Hsuan Yang;
403	Information Bottleneck: Exact Analysis of (Quantized) Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We investigate the information bottleneck in quantized neural networks, allowing us to compute the exact mutual information and provide an analysis free from estimation artifacts.	Stephan Sloth Lorenzen; Christian Igel; Mads Nielsen;
404	Information Gain Propagation: A New Way to Graph Active Learning with Soft Labels Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A new way to graph active learning with soft labels	Wentao Zhang; Yexin Wang; Zhenbang You; Meng Cao; Ping Huang; Jiulong Shan; Zhi Yang; Bin CUI;
405	Information Prioritization Through Empowerment in Visual Model-based RL Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Empowerment along with mutual information maximization helps learn functionally relevant factors in visual model-based RL	Homanga Bharadhwaj; Mohammad Babaeizadeh; Dumitru Erhan; Sergey Levine;
406	Interacting Contour Stochastic Gradient Langevin Dynamics Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose an interacting contour stochastic gradient Langevin dynamics sampler and prove it can be theoretically more efficient than a single-chain process with an equivalent computational budget.	Wei Deng; Siqi Liang; Botao Hao; Guang Lin; Faming Liang;
407	Interpretable Unsupervised Diversity Denoising and Artefact Removal Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work proposes a new architecture for unsupervised, interpretable and diverse image restoration while achieving state-of-the-art results on numerous commonly used benchmarks across multiple image domains.	Mangal Prakash; Mauricio Delbracio; Peyman Milanfar; Florian Jug;
408	IntSGD: Adaptive Floatless Compression of Stochastic Gradients Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the provably convergent and computationally cheap IntSGD algorithm for efficient distributed machine learning.	Konstantin Mishchenko; Bokun Wang; Dmitry Kovalev; Peter Richt�rik;
409	Invariant Causal Representation Learning for Out-of-Distribution Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose invariant Causal Representation Learning (iCaRL), an approach that enables out-of-distribution (OOD) generalization in the nonlinear setting (i.e., nonlinear representations and nonlinear classifiers).	Chaochao Lu; Yuhuai Wu; Jos� Miguel Hern�ndez-Lobato; Bernhard Sch�lkopf;
410	Inverse Online Learning: Understanding Non-Stationary and Reactionary Policies Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a practical algorithm for retrospectively estimating such perceived effects, alongside the process through which agents update them, using a novel architecture built upon an expressive family of deep state-space models.	Alex Chan; Alicia Curth; Mihaela van der Schaar;
411	Is Fairness Only Metric Deep? Evaluating and Addressing Subgroup Gaps in Deep Metric Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide a benchmark for fairness in the scope of deep metric learning; investigate fairness impacts of learned representations on downstream classification; and provide a novel method for reducing subgroup gaps in deep metric learning methods.	Natalie Dullerud; Karsten Roth; Kimia Hamidieh; Nicolas Papernot; Marzyeh Ghassemi;
412	Is High Variance Unavoidable in RL? A Case Study in Continuous Control Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: we study sources of variance in RL and propose methods to decrease it.	Johan Bjorck; Carla P Gomes; Kilian Q Weinberger;
413	Is Homophily A Necessity for Graph Neural Networks? Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Our work carefully characterizes these conditions and provides supporting theoretical understanding and empirical observations.	Yao Ma; Xiaorui Liu; Neil Shah; Jiliang Tang;
414	Is Importance Weighting Incompatible with Interpolating Classifiers? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We theoretically and empirically demonstrate that importance weighting can be effective in handling distribution shifts in overparameterized classifiers.	Ke Alexander Wang; Niladri Shekhar Chatterji; Saminul Haque; Tatsunori Hashimoto;
415	Iterated Reasoning with Mutual Information in Cooperative and Byzantine Decentralized Teaming Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a MARL framework with iterated and hierarchical rationalizability through mutual information for decision-making in fully-decentralized cooperative and Byzantine scenarios.	Sachin G Konan; Esmaeil Seraj; Matthew Gombolay;
416	Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-design Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new graph-based generative model for antibody design	Wengong Jin; Jeremy Wohlwend; Regina Barzilay; Tommi S. Jaakkola;
417	It Takes Four to Tango: Multiagent Self Play for Automatic Curriculum Generation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose Curriculum Self Play (CuSP), an automated goal generation framework that seeks to satisfy these desiderata by virtue of a multi-player game with 4 agents.	Yuqing Du; Pieter Abbeel; Aditya Grover;
418	It Takes Two to Tango: Mixup for Deep Metric Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We systematically study different mixup strategies in the context of deep metric learning, and study how to, and the effect of, mixing inputs, features, and output embeddings.	Shashanka Venkataramanan; Bill Psomas; Ewa Kijak; laurent amsaleg; Konstantinos Karantzalos; Yannis Avrithis;
419	Joint Shapley Values: A Measure of Joint Feature Importance Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a direct extension of Shapley’s value to sets of features, thus extending the Shapley value’s intuition: a set of feature’s average effect on a model’s prediction	Chris Harris; Richard Pymar; Colin Rowat;
420	Knowledge Infused Decoding Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new decoding algorithm for language model generation, to obtain better performance in knowledge-intensive tasks.	Ruibo Liu; Guoqing Zheng; Shashank Gupta; Radhika Gaonkar; Chongyang Gao; Soroush Vosoughi; Milad Shokouhi; Ahmed Hassan Awadallah;
421	Knowledge Removal in Sampling-based Bayesian Inference Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper proposes the first machine unlearning algorithm for MCMC.	Shaopeng Fu; Fengxiang He; Dacheng Tao;
422	L0-Sparse Canonical Correlation Analysis Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new $\ell_0$-CCA method for learning correlated representations based on sparse subsets of variables from two observed modalities.	Ofir Lindenbaum; Moshe Salhov; Amir Averbuch; Yuval Kluger;
423	Label Encoding for Regression Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose binary-encoded labels (BEL) which improves regression by generalizing the application of binary classification.	Deval Shah; Zi Yu Xue; Tor Aamodt;
424	Language-driven Semantic Segmentation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a language-driven approach that enables synthesis of zero-shot semantic segmentation models from arbitrary label sets at test time.	Boyi Li; Kilian Q Weinberger; Serge Belongie; Vladlen Koltun; Rene Ranftl;
425	Large Learning Rate Tames Homogeneity: Convergence and Balancing Effect Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Large learning rate well beyond 2/L provably induces an implicit regularization effect of balancing in gradient descent for matrix factorization.	Yuqing Wang; Minshuo Chen; Tuo Zhao; Molei Tao;
426	Large-Scale Representation Learning on Graphs Via Bootstrapping Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To address these challenges, we introduce Bootstrapped Graph Latents (BGRL) – a graph representation learning method that learns by predicting alternative augmentations of the input.	Shantanu Thakoor; Corentin Tallec; Mohammad Gheshlaghi Azar; Mehdi Azabou; Eva L Dyer; Remi Munos; Petar Velickovic; Michal Valko;
427	Latent Image Animator: Learning to Animate Image Via Latent Space Navigation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Image animation via latent space navigation	Yaohui Wang; Di Yang; Francois Bremond; Antitza Dantcheva;
428	Latent Variable Sequential Set Transformers for Joint Multi-Agent Motion Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: New Transformer-based architecture for socially consistent motion forecasting. Achieves SotA performance on NuScenes at a fraction of the compute of competing methods.	Roger Girgis; Florian Golemo; Felipe Codevilla; Martin Weiss; Jim Aldon D’Souza; Samira Ebrahimi Kahou; Felix Heide; Christopher Pal;
429	Learnability Lock: Authorized Learnability Control Through Adversarial Invertible Transformations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To tackle this issue, this paper introduces and investigates a new concept called ”learnability lock” for securing the process of data authorization.	Weiqi Peng; Jinghui Chen;
430	Learnability of Convolutional Neural Networks for Infinite Dimensional Input Via Mixed and Anisotropic Smoothness Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we investigate the approximation and estimation errors of the (dilated) convolutional neural networks when the input is infinite dimensional.	Sho Okumoto; Taiji Suzuki;
431	Learned Simulators for Turbulence Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learned simulators that outperform baselines in capturing turbulent dynamics at low resolution across multiple challenging turbulence domains.	Kim Stachenfeld; Drummond Buschman Fielding; Dmitrii Kochkov; Miles Cranmer; Tobias Pfaff; Jonathan Godwin; Can Cui; Shirley Ho; Peter Battaglia; Alvaro Sanchez-Gonzalez;
432	Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a method of processing the 3D torsion angles of a molecular conformer to learn tetrahedral chirality while integrating a novel invariance to rotations about internal molecular bonds directly into the model architecture.	Keir Adams; Lagnajit Pattanaik; Connor W. Coley;
433	Learning Altruistic Behaviours in Reinforcement Learning Without External Rewards Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose and investigate unsupervised training of agents to behave altruistically towards others by actively maximizing others’ choice.	Tim Franzmeyer; Mateusz Malinowski; Joao F. Henriques;
434	Learning A Subspace of Policies for Online Adaptation in Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose an approach to learn a subspace of policies that are robust to different variations of the train environment.	Jean-Baptiste Gaya; Laure Soulier; Ludovic Denoyer;
435	Learning Audio-Visual Speech Representation By Masked Multimodal Cluster Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A self-supervised learning framework for audio-visual speech data, which uses only 30h of labeled data to match the SOTA lip-reading model trained on 31k hours of data (34.6% vs 33.6% WER), and further outperforms the SOTA with 70x less data (30.6%).	Bowen Shi; Wei-Ning Hsu; Kushal Lakhotia; Abdelrahman Mohamed;
436	Learning-Augmented $k$-means Clustering Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study the $k$-means problem augmented with a learning-based predictor that gives noisy information about true labels.	Jon Ergun; Zhili Feng; Sandeep Silwal; David Woodruff; Samson Zhou;
437	Learning By Directional Gradient Descent Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Computing directional derivative of a recurrent function along a candidate direction, and using it to create a valid descent direction.	David Silver; Anirudh Goyal; Ivo Danihelka; Matteo Hessel; Hado van Hasselt;
438	Learning Causal Relationships from Conditional Moment Restrictions By Importance Weighting Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learning causal relationships under conditional moment restrictions by importance weighting using the conditional density ratio function.	Masahiro Kato; Masaaki Imaizumi; Kenichiro McAlinn; Shota Yasui; Haruo Kakehi;
439	Learning Continuous Environment Fields Via Implicit Functions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel scene representation that can dynamically change behaviors of agents inside the scene.	Xueting Li; Sifei Liu; Shalini De Mello; Xiaolong Wang; Ming-Hsuan Yang; Jan Kautz;
440	Learning Curves for Continual Learning in Neural Networks: Self-knowledge Transfer and Forgetting Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We analyze the generalization performance of continual learning in the NTK regime and identify key properties of knowledge transfer and forgetting.	Ryo Karakida; Shotaro Akaho;
441	Learning Curves for Gaussian Process Regression with Power-Law Priors and Targets Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We derive the power-law decay rate of the generalization error in Gaussian process regression depending on the eigenspectrum of the prior and the target.	Hui Jin; Pradeep Kr. Banerjee; Guido Montufar;
442	Learning Curves for SGD on Structured Features Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: The average case test risk for stochastic gradient descent on mean square error is computed in terms of feature covariance structure.	Blake Bordelon; Cengiz Pehlevan;
443	Learning Discrete Structured Variational Auto-Encoder Using Natural Evolution Strategies Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we use Natural Evolution Strategies (NES), a class of gradient-free black-box optimization algorithms, to learn discrete structured VAEs.	Alon Berliner; Guy Rotman; Yossi Adi; Roi Reichart; Tamir Hazan;
444	Learning Disentangled Representation By Exploiting Pretrained Generative Models: A Contrastive Learning View Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: DisCo is a new contrastive learning framework to leverage pretrained generative models to jointly learn disentangled representation and discover disentangled directions in the latent space.	Xuanchi Ren; Tao Yang; Yuwang Wang; Wenjun Zeng;
445	Learning Distributionally Robust Models at Scale Via Composite Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we show how different variants of DRO are simply instances of a finite-sum composite optimization for which we provide scalable methods.	Farzin Haddadpour; Mohammad Mahdi Kamani; Mehrdad Mahdavi; amin karbasi;
446	Learning Efficient Image Super-Resolution Networks Via Structure-Regularized Pruning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learning efficient compressed models for bother lightweight and large image super-resolution networks	Yulun Zhang; Huan Wang; Can Qin; Yun Fu;
447	Learning Efficient Online 3D Bin Packing on Packing Configuration Trees Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose to enhance the practical applicability of online 3D-BPP via learning on a hierarchical packing configuration tree which makes the DRL model easy to deal with practical constraints and well-performing even with continuous solution space.	Hang Zhao; Yang Yu; Kai Xu;
448	Learning Features with Parameter-Free Layers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper introduces a new design paradigm rethinking a parameter-free operation as the main building block of network architecture.	Dongyoon Han; YoungJoon Yoo; Beomyoung Kim; Byeongho Heo;
449	Learning Generalizable Representations for Reinforcement Learning Via Adaptive Meta-learner of Behavioral Similarities Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we propose a novel meta-learner-based framework for representation learning regarding behavioral similarities in reinforcement learning.	Jianda Chen; Sinno Pan;
450	Learning Graphon Mean Field Games and Approximate Nash Equilibria Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose, analyze and solve a novel, theoretically well-founded graph-based mean field game for Nash equilibria in discrete-time dynamical systems on otherwise infeasibly large dense graphs.	Kai Cui; Heinz Koeppl;
451	LEARNING GUARANTEES FOR GRAPH CONVOLUTIONAL NETWORKS ON THE STOCHASTIC BLOCK MODEL Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper we present the first provable guarantees for one of the best-studied families of graph neural network models, Graph Convolutional Networks (GCNs), for semi- supervised community detection tasks.	Wei Lu;
452	Learning Hierarchical Structures with Differentiable Nondeterministic Stacks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a new stack-augmented RNN with strong results on CFL language modeling tasks.	Brian DuSell; David Chiang;
453	Learning Invariant Representations on Multilingual Language Models for Unsupervised Cross-Lingual Transfer Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose an importance-weighted domain adaptation method for unsupervised cross-lingual learning that is effective at correcting class prior shifts, a distributional property that negatively affects transfer performance.	Ruicheng Xian; Heng Ji; Han Zhao;
454	Learning Long-Term Reward Redistribution Via Randomized Return Decomposition Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose randomized return decomposition, a novel reward redistribution algorithm, which establishes a surrogate optimization problem to scale up learning in long-horizon tasks.	Zhizhou Ren; Ruihan Guo; Yuan Zhou; Jian Peng;
455	Learning Meta-features for AutoML Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel approach to learn dataset meta-features for AutoML.	Herilalaina Rakotoarison; Louisot Milijaona; Andry RASOANAIVO; Michele Sebag; Marc Schoenauer;
456	Learning More Skills Through Optimistic Exploration Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learn more skills by adding an information gain exploration bonus based on discriminator ensemble disagreement.	DJ Strouse; Kate Baumli; David Warde-Farley; Volodymyr Mnih; Steven Stenberg Hansen;
457	Learning Multimodal VAEs Through Mutual Supervision Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Here we re-purpose semi-supervised VAEs to leverage mutual supervision between encoding distributions, allowing us to learn multi-modal VAEs with partially obsereved data.	Tom Joy; Yuge Shi; Philip Torr; Tom Rainforth; Sebastian M Schmon; Siddharth N;
458	Learning Neural Contextual Bandits Through Perturbed Rewards Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose to perturb the rewards when updating the neural network to eliminate the need of explicit exploration and the corresponding computational overhead.	Yiling Jia; Weitong ZHANG; Dongruo Zhou; Quanquan Gu; Hongning Wang;
459	Learning Object-Oriented Dynamics for Planning from Text Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose an Object-Oriented Text Dynamic (OOTD) model for solving decision-making problems in the text domain.	Guiliang Liu; Ashutosh Adhikari; Amir-massoud Farahmand; Pascal Poupart;
460	Learning Optimal Conformal Classifiers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Conformal training allows to train classifier and conformal predictor end-to-end, optimizing average confidence set size (inefficiency) or other application-specific losses defined on confidence sets.	David Stutz; Krishnamurthy Dj Dvijotham; Ali Taylan Cemgil; Arnaud Doucet;
461	Learning Prototype-oriented Set Representations for Meta-Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A plug-and-play framework for set-structured input tasks	Dan dan Guo; Long Tian; Minghe Zhang; Mingyuan Zhou; Hongyuan Zha;
462	Learning Pruning-Friendly Networks Via Frank-Wolfe: One-Shot, Any-Sparsity, And No Retraining Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel and state-of-the-art one-shot pruning method, which can generate sparse networks at any pruning ratio in one pruning and without any retraining.	Lu Miao; Xiaolong Luo; Tianlong Chen; Wuyang Chen; Dong Liu; Zhangyang Wang;
463	Learning Representation from Neural Fisher Kernel with Low-rank Approximation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we study the representation of neural networks from the view of kernels.	Ruixiang ZHANG; Shuangfei Zhai; Etai Littwin; Joshua M. Susskind;
464	Learning Scenario Representation for Solving Two-stage Stochastic Integer Programs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper provides a CVAE based method to learn scenario representations for solving stochastic integer programs.	Yaoxin Wu; Wen Song; Zhiguang Cao; Jie Zhang;
465	Learning State Representations Via Retracing in Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce Learning via Retracing, a novel self-supervised framework based on temporal cycle-consistency assumption of the transition dynamics, for improved learning of the representation (and the dynamics model) in RL tasks.	Changmin Yu; Dong Li; Jianye HAO; Jun Wang; Neil Burgess;
466	Learning Strides in Convolutional Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce DiffStride, the first downsampling layer with learnable strides for convolutional neural networks.	Rachid Riad; Olivier Teboul; David Grangier; Neil Zeghidour;
467	Learning Synthetic Environments and Reward Networks for Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose an evolution-based approach to meta-learn synthetic neural environments and reward neural networks for reinforcement learning.	Fabio Ferreira; Thomas Nierhoff; Andreas S�linger; Frank Hutter;
468	Learning Temporally Latent Causal Processes from General Temporal Data Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Propose two provable conditions and training framework with which temporally latent causal processes are identifiable from observed variables.	Weiran Yao; Yuewen Sun; Alex Ho; Changyin Sun; Kun Zhang;
469	Learning The Dynamics of Physical Systems from Sparse Observations with Finite Element Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A continuous-time graph neural network model for spatio-temporal forecasting that can structurally incorporate prior knowledge	Marten Lienen; Stephan G�nnemann;
470	Learning to Annotate Part Segmentation with Gradient Matching Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a gradient-matching-based method to learn annotator which is able to label generated images with part segmentation by decoding the generator features into segmentation masks.	Yu Yang; Xiaotian Cheng; Hakan Bilen; Xiangyang Ji;
471	Learning to Dequantise with Truncated Flows Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learning a variational dequantisation scheme with truncated/bounded-support distributions	Shawn Tan; Chin-Wei Huang; Alessandro Sordoni; Aaron Courville;
472	Learning to Downsample for Segmentation of Ultra-High Resolution Images Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a method for learning to downsample ultra high-resolution images that reflects the importance of each location.	Chen Jin; Ryutaro Tanno; Thomy Mertzanidou; Eleftheria Panagiotaki; Daniel C. Alexander;
473	Learning to Extend Molecular Scaffolds with Structural Motifs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new fragment-based generative model of molecules that can be constrained to include an arbitrary subgraph (scaffold).	Krzysztof Maziarz; Henry Richard Jackson-Flux; Pashmina Cameron; Finton Sirockin; Nadine Schneider; Nikolaus Stiefl; Marwin Segler; Marc Brockschmidt;
474	Learning to Generalize Across Domains on Single Test Samples Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We leverage a meta-learning paradigm to learn our model to acquire the ability of adaptation with single samples at training time so as to further adapt itself to each single test sample at test time.	Zehao Xiao; Xiantong Zhen; Ling Shao; Cees G. M. Snoek;
475	Learning to Map for Active Semantic Goal Navigation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A framework for object goal navigation that actively learns to predict semantic maps and choose long-term goals based on uncertainty measures.	Georgios Georgakis; Bernadette Bucher; Karl Schmeckpeper; Siddharth Singh; Kostas Daniilidis;
476	Learning to Remember Patterns: Pattern Matching Memory Networks for Traffic Forecasting Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We presents a novel graph convolutional memory network, PM-MemNet, for the traffic forecasting.	Hyunwook Lee; Seungmin Jin; Hyeshin Chu; Hongkyu Lim; Sungahn Ko;
477	Learning to Schedule Learning Rate with Graph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel Graph-Network-based Scheduler (GNS), which is both informative to to encode rich information and generalizable to different architectures.	Yuanhao Xiong; Li-Cheng Lan; Xiangning Chen; Ruochen Wang; Cho-Jui Hsieh;
478	Learning Towards The Largest Margins Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we attempt to address this limitation by formulating the principled optimization objective as learning towards the largest margins.	Xiong Zhou; Xianming Liu; Deming Zhai; Junjun Jiang; Xin Gao; Xiangyang Ji;
479	Learning Transferable Motor Skills with Hierarchical Latent Mixture Policies Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: An approach to learn reusable and transferable skills from data via a hierarchical latent mixture policy, which can significantly improve sample efficiency and asymptotic performance on downstream RL tasks	Dushyant Rao; Fereshteh Sadeghi; Leonard Hasenclever; Markus Wulfmeier; Martina Zambelli; Giulia Vezzani; Dhruva Tirumala; Yusuf Aytar; Josh Merel; Nicolas Heess; raia hadsell;
480	Learning Transferable Reward for Query Object Localization with Policy Adaptation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a reinforcement learning based approach to the problem of query object localization, where an agent is trained to localize objects of interest specified by a small exemplary set.	Tingfeng Li; Shaobo Han; Martin Renqiang Min; Dimitris N. Metaxas;
481	Learning Value Functions from Undirected State-only Experience Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a method for offline value learning without action labels.	Matthew Chang; Arjun Gupta; Saurabh Gupta;
482	Learning Versatile Neural Architectures By Propagating Network Codes Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: An efficient NAS method by inverting neural predictors to directly update architectures and a multitask NAS benchmark for cross-task architecture design and analysis. We first introduce a unified design space for multiple tasks and build a multitask NAS benchmark (NAS-Bench-MR) on many wildly used datasets, including ImageNet, Cityscapes, KITTI, and HMDB51.	Mingyu Ding; Yuqi Huo; Haoyu Lu; Linjie Yang; Zhe Wang; Zhiwu Lu; Jingdong Wang; Ping Luo;
483	Learning Vision-Guided Quadrupedal Locomotion End-to-End with Cross-Modal Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a novel end-to-end Reinforcement Learning approach called LocoTransformer, leveraging both visual inputs and proprioceptive states, for locomotion control in both simulation and with real robots.	Ruihan Yang; Minghao Zhang; Nicklas Hansen; Huazhe Xu; Xiaolong Wang;
484	Learning Weakly-supervised Contrastive Representations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a weakly-supervised contrastive learning framework that considers auxiliary information (additional sources of information from data).	Yao-Hung Hubert Tsai; Tianqin Li; Weixin Liu; Peiyuan Liao; Ruslan Salakhutdinov; Louis-Philippe Morency;
485	Learning with Noisy Labels Revisited: A Study Using Real-World Human Annotations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we revisit the problem of learning from noisy labels using human annotated CIFAR datasets we collected from Amazon Mechanical Turks.	Jiaheng Wei; Zhaowei Zhu; Hao Cheng; Tongliang Liu; Gang Niu; Yang Liu;
486	Learn Locally, Correct Globally: A Distributed Algorithm for Training Graph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose LLCG a communication efficient distributed algorithm for training GNNs.	Morteza Ramezani; Weilin Cong; Mehrdad Mahdavi; Mahmut Kandemir; Anand Sivasubramaniam;
487	Leveraging Automated Unit Tests for Unsupervised Code Translation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We leverage automatically created multilingual unit tests to improve unsupervised machine translation methods for source code and substantially outperform the state-of-the-art on all the language pairs we consider.	Baptiste Roziere; Jie Zhang; Francois Charton; Mark Harman; Gabriel Synnaeve; Guillaume Lample;
488	Leveraging Unlabeled Data to Predict Out-of-distribution Performance Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we investigate methods for predicting the target domain accuracy using only labeled source data and unlabeled target data.	Saurabh Garg; Sivaraman Balakrishnan; Zachary Chase Lipton; Behnam Neyshabur; Hanie Sedghi;
489	LFPT5: A Unified Framework for Lifelong Few-shot Language Learning Based on Prompt Tuning of T5 Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We define a challenging yet practical problem as Lifelong Few-shot Language Learning and propose a unified framework for it based on prompt tuning of T5.	Chengwei Qin; Shafiq Joty;
490	LIGS: Learnable Intrinsic-Reward Generation Selection for Multi-Agent Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we introduce a new general framework for improving coordination and performance of multi-agent reinforcement learners (MARL).	David Henry Mguni; Taher Jafferjee; Jianhong Wang; Nicolas Perez-Nieves; Oliver Slumbers; Feifei Tong; Yang Li; Jiangcheng Zhu; Yaodong Yang; Jun Wang;
491	Likelihood Training of Schr�dinger Bridge Using Forward-Backward SDEs Theory Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a new computational framework, grounded on Forward-Backward SDEs theory, for the log-likelihood training of Schr?dinger Bridge and provide theoretical connections to score-baesd generative models.	Tianrong Chen; Guan-Horng Liu; Evangelos Theodorou;
492	Linking Emergent and Natural Languages Via Corpus Transfer Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We find that pre-training on an emergent language corpus improves natural language tasks in a low resource setup, and propose a metric to predict such a transferability.	Shunyu Yao; Mo Yu; Yang Zhang; Karthik R Narasimhan; Joshua B. Tenenbaum; Chuang Gan;
493	Lipschitz-constrained Unsupervised Skill Discovery Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Lipschitz-constrained Skill Discovery (LSD), which encourages the agent to discover more dynamic and diverse skills without external rewards and additional prior knowledge, enabling zero-shot control on goal-reaching downstream tasks.	Seohong Park; Jongwook Choi; Jaekyeom Kim; Honglak Lee; Gunhee Kim;
494	Local Feature Swapping for Generalization in Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a simple yet effective layer increasing the generalization abilities of reinforcement learning agents	David Bertoin; Emmanuel Rachelson;
495	Long Expressive Memory for Sequence Modeling Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A novel method for sequence modeling based on multiscale ODEs that is provably able to learn very long-term dependencies while being sufficiently expressive to outperform state-of-the-art recurrent sequence models.	T. Konstantin Rusch; Siddhartha Mishra; N. Benjamin Erichson; Michael W. Mahoney;
496	Looking Back on Learned Experiences For Class/task Incremental Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we shed light on an on-call transfer set to provide past experiences whenever a new task arises in the data stream.	Mozhgan PourKeshavarzi; Guoying Zhao; Mohammad Sabokrou;
497	LoRA: Low-Rank Adaptation of Large Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Finetuning updates have a low intrinsic rank which allows us to train only the rank decomposition matrices of certain weights, yielding better performance and practical benefits.	Edward J Hu; yelong shen; Phillip Wallis; Zeyuan Allen-Zhu; Yuanzhi Li; Shean Wang; Lu Wang; Weizhu Chen;
498	LORD: Lower-Dimensional Embedding of Log-Signature in Neural Rough Differential Equations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We reduce the complexity of processing higher depth log-signatures in NRDE.	JAEHOON LEE; Jinsung Jeon; Sheo yon Jhin; Jihyeon Hyeong; Jayoung Kim; Minju Jo; Kook Seungji; Noseong Park;
499	Lossless Compression with Probabilistic Circuits Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To overcome such problems, we establish a new class of tractable lossless compression models that permit efficient encoding and decoding: Probabilistic Circuits (PCs).	Anji Liu; Stephan Mandt; Guy Van den Broeck;
500	Low-Budget Active Learning Via Wasserstein Distance: An Integer Programming Approach Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose an integer optimization problem for active learning and demonstrate how optimally selecting which points to label can significantly improve classifiers under low labeling budgets.	Rafid Mahmood; Sanja Fidler; Marc T Law;
501	MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retraining Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a differential-geometry-based technique to provably sample uniformly from the data manifold of a trained Deep Generative Network without the need for retraining.	Ahmed Imtiaz Humayun; Randall Balestriero; Richard Baraniuk;
502	MAML Is A Noisy Contrastive Learner Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The Model-agnostic meta learning (MAML) algorithm is a noisy supervised contrastive learner where the noise comes from random initialization and cross-task interference.	Chia Hsiang Kao; Wei-Chen Chiu; Pin-Yu Chen;
503	Map Induction: Compositional Spatial Submap Learning for Efficient Exploration in Novel Environments Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Modelling Map Induction for efficient exploration in novel environments.	Sugandha Sharma; Aidan Curtis; Marta Kryven; Joshua B. Tenenbaum; Ila R Fiete;
504	Mapping Conditional Distributions for Domain Adaptation Under Generalized Target Shift Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel theoretically grounded approach for domain adaptation under Generalized Target Shift; it learns a map between pretrained source and target representations that matches conditional distributions and recovers target proportions.	Matthieu Kirchmeyer; Alain Rakotomamonjy; Emmanuel de Bezenac; patrick gallinari;
505	Mapping Language Models to Grounded Conceptual Spaces Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Mapping text-only pre-trained language models to grounded conceptual worlds.	Roma Patel; Ellie Pavlick;
506	Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We proposed a model-free off-policy algorithm for image-based continuous control that significantly outperforms previous methods both in sample and time complexity.	Denis Yarats; Rob Fergus; Alessandro Lazaric; Lerrel Pinto;
507	$\mathrm{SO}(2)$-Equivariant Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes equivariant DQN and equivariant SAC that significantly improve the sample efficiency of RL in robotic manipulation.	Dian Wang; Robin Walters; Robert Platt;
508	Maximizing Ensemble Diversity in Deep Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Maximizing diversity in neural network improves performance ensemble based reinforcement learning	Hassam Sheikh; Mariano Phielipp; Ladislau Boloni;
509	Maximum Entropy RL (Provably) Solves Some Robust RL Problems Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Maximum Entropy RL (provably) solves some robust RL problems.	Benjamin Eysenbach; Sergey Levine;
510	Maximum N-times Coverage for Vaccine Design Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce the maximum $n$-times coverage problem that selects $k$ overlays to maximize the summed coverage of weighted elements, where each element must be covered at least $n$ times, and show its importance for vaccine design. We introduce the maximum n-times coverage problem that selects $k$ overlays to maximize the summed coverage of weighted elements, where each element must be covered at least $n$ times.	Ge Liu; Alexander Dimitrakakis; Brandon Carter; David Gifford;
511	MCMC Should Mix: Learning Energy-Based Model with Flow-Based Backbone Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learning energy-based models with mixing Markov chains.	Erik Nijkamp; Ruiqi Gao; Pavel Sountsov; Srinivas Vasudevan; Bo Pang; Song-Chun Zhu; Ying Nian Wu;
512	Measuring CLEVRness: Black-box Testing of Visual Reasoning Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Black box testing of visual reasoning models as a two-player game.	Spyridon Mouselinos; Henryk Michalewski; Mateusz Malinowski;
513	Measuring The Interpretability of Unsupervised Representations Via Quantized Reversed Probing Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose quantized reverse probing as a information-theoretic measure to assess the degree to which self-supervised visual representations align with human-interpretable concepts.	Iro Laina; Yuki M Asano; Andrea Vedaldi;
514	Memorizing Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose to use an external memory module to allow instant utilization of newly acquired knowledge.	Yuhuai Wu; Markus Norman Rabe; DeLesley Hutchins; Christian Szegedy;
515	Memory Augmented Optimizers for Deep Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a framework of memory augmented optimizers and empirically show that the class of optimizers provide accelerated convergence and even better test performance. We show the proposed optimizers converge in smooth strongly convex setting.	Paul-Aymeric Martin McRae; Prasanna Parthasarathi; Mido Assran; Sarath Chandar;
516	Memory Replay with Data Compression for Continual Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose memory replay with data compression, which is an important yet neglected baseline and a promising direction for continual learning.	Liyuan Wang; Xingxing Zhang; Kuo Yang; Longhui Yu; Chongxuan Li; Lanqing HONG; Shifeng Zhang; Zhenguo Li; Yi Zhong; Jun Zhu;
517	Mention Memory: Incorporating Textual Knowledge Into Transformers Through Entity Mention Attention Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Incorporate information from text corpus into Transformer model through within-model attention over table of entity mention representations.	Michiel de Jong; Yury Zemlyanskiy; Nicholas FitzGerald; Fei Sha; William W. Cohen;
518	Message Passing Neural PDE Solvers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper introduces a message passing neural PDE solver that replaces all heuristically designed components in numerical PDE solvers with backprop-optimized neural function approximators.	Johannes Brandstetter; Daniel E. Worrall; Max Welling;
519	Meta Discovery: Learning to Discover Novel Classes Given Very Limited Data Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we demystify assumptions behind L2DNC and find that high-level semantic features should be shared among the seen and unseen classes.	Haoang Chi; Feng Liu; Wenjing Yang; Long Lan; Tongliang Liu; Bo Han; Gang Niu; Mingyuan Zhou; Masashi Sugiyama;
520	Meta-Imitation Learning By Watching Video Demonstrations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present an approach of meta-imitation learning by watching video demonstrations from humans.	Jiayi Li; Tao Lu; Xiaoge Cao; Yinghao Cai; Shuo Wang;
521	Meta Learning Low Rank Covariance Factors for Energy Based Deterministic Uncertainty Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel meta learning algorithm which learns low rank covariance factors, and utilizes an energy-based inference to achieve a calibrated prediction.	Jeffrey Ryan Willette; Hae Beom Lee; Juho Lee; Sung Ju Hwang;
522	MetaMorph: Learning Universal Controllers with Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We learn a transformer based general purpose controller for a modular robot design space which can zero-shot generalize to unseen variations in dynamics, kinematics, new morphologies and tasks.	Agrim Gupta; Linxi Fan; Surya Ganguli; Li Fei-Fei;
523	MetaShift: A Dataset of Datasets for Evaluating Contextual Distribution Shifts and Training Conflicts Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We leverage annotated subsets within a heterogeneous dataset to evaluate the performance of learning algorithms to distribution shifts and to visualize training dynamics.	Weixin Liang; James Zou;
524	Mind The Gap: Domain Gap Control for Single Shot Domain Adaptation for Generative Adversarial Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose several regularizers to control the domain transfer for single shot domain adaptation in the context of generative adversarial networks.	Peihao Zhu; Rameen Abdal; John Femiani; Peter Wonka;
525	MiniF2F: A Cross-system Benchmark for Formal Olympiad-level Mathematics Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present $\textsf{miniF2F}$, a dataset of formal Olympiad-level mathematics problems statements intended to provide a unified cross-system benchmark for neural theorem proving.	Kunhao Zheng; Jesse Michael Han; Stanislas Polu;
526	Minimax Optimality (Probably) Doesn’t Imply Distribution Learning for GANs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Under the standard crypto assumption that local pseudorandom generators exist, we show that even a global optimizer for the population WGAN objective need not be close to the true distribution.	Sitan Chen; Jerry Li; Yuanzhi Li; Raghu Meka;
527	Minimax Optimization with Smooth Algorithmic Adversaries Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a tractable formulation of minimax optimization by modeling the adversary’s algorithm, and present new algorithms which are guaranteed to converge and find appropriate stationary points.	Tanner Fiez; Chi Jin; Praneeth Netrapalli; Lillian J Ratliff;
528	Mirror Descent Policy Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A theory-grounded practical algorithm for policy optimization in RL, which is conceptually simpler and performs better or on par to SOTA.	Manan Tomar; Lior Shani; Yonathan Efroni; Mohammad Ghavamzadeh;
529	Missingness Bias in Model Debugging Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We investigate how current missingness approximations for model debugging can impose undesirable biases on the model predictions and hinder our ability to debug models, and we show how transformer-based architectures can side-step these issues.	Saachi Jain; Hadi Salman; Eric Wong; Pengchuan Zhang; Vibhav Vineet; Sai Vemprala; Aleksander Madry;
530	MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Light-weight and general-purpose vision transformers for mobile devices	Sachin Mehta; Mohammad Rastegari;
531	Model Agnostic Interpretability for Multiple Instance Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose and compare several methods for model-agnostic interpretability for multiple instance learning.	Joseph Early; Christine Evers; SArvapali Ramchurn;
532	Model-augmented Prioritized Experience Replay Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel experience replay which employs additional auxiliary learnable features as well as TD-errors for prioritizing experiences	Youngmin Oh; Jinwoo Shin; Eunho Yang; Sung Ju Hwang;
533	Model-Based Offline Meta-Reinforcement Learning with Regularization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes a novel offline Meta-RL algorithm with regularization, which has provable performance improvement and outperforms the existing baselines empirically.	Sen Lin; Jialin Wan; Tengyu Xu; Yingbin Liang; Junshan Zhang;
534	Modeling Label Space Interactions in Multi-label Classification Using Box Embeddings Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Improving the consistency for multi-label classification by modeling label space interactions using Box Embeddings.	Dhruvesh Patel; Pavitra Dangati; Jay-Yoon Lee; Michael Boratko; Andrew McCallum;
535	Model Zoo: A Growing Brain That Learns Continually Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Continual learning methods can benefit by splitting the capacity of the learner and we leverage this in our method Model Zoo, which demonstrates large gains in accuracy on a variety of continual learning benchmarks.	Rahul Ramesh; Pratik Chaudhari;
536	Modular Lifelong Reinforcement Learning Via Neural Composition Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We explore the problem of lifelong RL of functionally composable knowledge, and develop an algorithm that demonstrates zero-shot and forward transfer, avoidance of forgetting, and backward transfer in discrete 2-D and robotic manipulation domains.	Jorge A Mendez; Harm van Seijen; ERIC EATON;
537	MonoDistill: Learning Spatial Features for Monocular 3D Object Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the MonoDistill, which introduces spatial cues to the monocular 3D detector based on the knowledge distillation mechanism.	Zhiyu Chong; Xinzhu Ma; Hong Zhang; Yuxin Yue; Haojie Li; Zhihui Wang; Wanli Ouyang;
538	Monotonic Differentiable Sorting Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To address this issue, we propose a novel relaxation of conditional swap operations that guarantees monotonicity in differentiable sorting networks.	Felix Petersen; Christian Borgelt; Hilde Kuehne; Oliver Deussen;
539	MoReL: Multi-omics Relational Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel deep Bayesian generative model to efficiently infer a multi-partite graph encoding molecular interactions across such heterogeneous views, using a fused Gromov-Wasserstein (FGW) regularization between latent representations of corresponding views for integrative analysis.	Arman Hasanzadeh; Ehsan Hajiramezanali; Nick Duffield; Xiaoning Qian;
540	MT3: Multi-Task Multitrack Music Transcription Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Unified framework for music transcription, jointly training a single model on six multi-instrument datasets and establishing a new SOTA for low-resource music transcription.	Joshua P Gardner; Ian Simon; Ethan Manilow; Curtis Hawthorne; Jesse Engel;
541	Multi-Agent MDP Homomorphic Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce globally equivariant multi-agent policy networks with distributed execution.	Elise van der Pol; Herke van Hoof; Frans A Oliehoek; Max Welling;
542	Multimeasurement Generative Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We formally map the problem of sampling from an unknown distribution with density $p_X$ in $\mathbb{R}^d$ to the problem of learning and sampling $p_\mathbf{Y}$ in $\mathbb{R}^{Md}$ obtained by convolving $p_X$ with a fixed factorial kernel: $p_\mathbf{Y}$ is referred to as M-density and the factorial kernel as multimeasurement noise model (MNM).	Saeed Saremi; Rupesh Kumar Srivastava;
543	Multi-Mode Deep Matrix and Tensor Factorization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper presents a framework of multi-mode deep matrix and tensor factorizations to explore and exploit the full nonlinearity of the data in matrices and tensors.	Jicong Fan;
544	Multi-objective Optimization By Learning Space Partition Related Papers Related Patents Related Grants Related Orgs Related Experts View Abstract: In contrast to single-objective optimization (SOO), multi-objective optimization (MOO) requires an optimizer to find the Pareto frontier, a subset of feasible solutions that are …	Yiyang Zhao; Linnan Wang; Kevin Yang; Tianjun Zhang; Tian Guo; Yuandong Tian;
545	Multiset-Equivariant Set Prediction with Approximate Implicit Differentiation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a better permutation-equivariance property for multisets and improve an existing set predictor that has this property with approximate implicit differentiation	Yan Zhang; David W Zhang; Simon Lacoste-Julien; Gertjan J. Burghouts; Cees G. M. Snoek;
546	Multi-Stage Episodic Control for Strategic Exploration in Text Games Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a multi-stage approach to playing text games that improves the score on Zork1 from around 40 to 103.	Jens Tuyls; Shunyu Yao; Sham M. Kakade; Karthik R Narasimhan;
547	Multi-Task Processes Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new family of stochastic processes that can infer multiple heterogeneous functions jointly given a few incomplete observations (i.e., some functions may not be observed at each input).	Donggyun Kim; Seongwoong Cho; Wonkwang Lee; Seunghoon Hong;
548	Multitask Prompted Training Enables Zero-Shot Task Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Can zero-shot generalization instead be directly induced by explicit multitask learning? To test this question at scale, we develop a system for easily mapping general natural language tasks into a human-readable prompted form.	Victor Sanh; Albert Webson; Colin Raffel; Stephen Bach; Lintang Sutawika; Zaid Alyafeai; Antoine Chaffin; Arnaud Stiegler; Arun Raja; Manan Dey; M Saiful Bari; Canwen Xu; Urmish Thakker; Shanya Sharma Sharma; Eliza Szczechla; Taewoon Kim; Gunjan Chhablani; Nihal Nayak; Debajyoti Datta; Jonathan Chang; Mike Tian-Jian Jiang; Han Wang; Matteo Manica; Sheng Shen; Zheng Xin Yong; Harshit Pandey; Rachel Bawden; Thomas Wang; Trishala Neeraj; Jos Rozen; Abheesht Sharma; Andrea Santilli; Thibault Fevry; Jason Alan Fries; Ryan Teehan; Teven Le Scao; Stella Biderman; Leo Gao; Thomas Wolf; Alexander M Rush;
549	NAS-Bench-Suite: NAS Evaluation Is (Now) Surprisingly Easy Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show that you cannot get away only with NAS-Bench-101 and -201; to fix this, we release a unified NAS benchmark suite with 25 benchmarks.	Yash Mehta; Colin White; Arber Zela; Arjun Krishnakumar; Guri Zabergja; Shakiba Moradian; Mahmoud Safari; Kaicheng Yu; Frank Hutter;
550	NASI: Label- and Data-agnostic Neural Architecture Search at Initialization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To overcome this limitation, we propose a novel NAS algorithm called NAS at Initialization (NASI) that exploits the capability of a Neural Tangent Kernel in being able to characterize the performance of candidate architectures at initialization, hence allowing model training to be completely avoided to boost the search efficiency.	Yao Shu; Shaofeng Cai; Zhongxiang Dai; Beng Chin Ooi; Bryan Kian Hsiang Low;
551	NASPY: Automated Extraction of Automated Machine Learning Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present NASPY, an end-to-end adversarial framework to extract the networkarchitecture of deep learning models from Neural Architecture Search (NAS).	Xiaoxuan Lou; Shangwei Guo; Jiwei Li; Yaoxin Wu; Tianwei Zhang;
552	NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict Aware Supernet Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: we identify one key issue of ViT supernet training that the supernet gradients and the sub-network gradients are likely to disagree with each other, and propose gradient conflict aware training.	Chengyue Gong; Dilin Wang; Meng Li; Xinlei Chen; Zhicheng Yan; Yuandong Tian; qiang liu; Vikas Chandra;
553	Natural Posterior Network: Deep Bayesian Predictive Uncertainty for Exponential Family Distributions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we propose the Natural Posterior Network (NatPN) for fast and high-quality uncertainty estimation for any task where the target distribution belongs to the exponential family.	Bertrand Charpentier; Oliver Borchert; Daniel Z�gner; Simon Geisler; Stephan G�nnemann;
554	Near-optimal Offline Reinforcement Learning with Linear Representation: Leveraging Variance Information with Pessimism Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Towards this goal, we study the statistical limits of offline reinforcement learning with linear model representations.	Ming Yin; Yaqi Duan; Mengdi Wang; Yu-Xiang Wang;
555	Near-Optimal Reward-Free Exploration for Linear Mixture MDPs with Plug-in Solver Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose near-optimal exploration algorithms for reward-free exploration with plug-in solver.	Xiaoyu Chen; Jiachen Hu; Lin Yang; Liwei Wang;
556	Network Augmentation for Tiny Deep Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: a training method for tiny neural networks	Han Cai; Chuang Gan; Ji Lin; song han;
557	NETWORK INSENSITIVITY TO PARAMETER NOISE VIA PARAMETER ATTACK DURING TRAINING Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We flatten the weight loss-landscape by introducing a parameter attack term in the loss function and demonstrate improved network insensitivity to noise common in analog neuromorphic hardware.	Julian B�chel; Fynn Firouz Faber; Dylan Richard Muir;
558	NeuPL: Neural Population Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose NeuPL, a general and efficient population learning framework that learns and represents diverse policies in symmetric zero-sum games within a single conditional network via self-play.	Siqi Liu; Luke Marris; Daniel Hennes; Josh Merel; Nicolas Heess; Thore Graepel;
559	Neural Contextual Bandits with Deep Representation and Shallow Exploration Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A new neural network based algorithm for contextual bandit problems with theoretical guarantees and empirical advantages.	Pan Xu; Zheng Wen; Handong Zhao; Quanquan Gu;
560	Neural Deep Equilibrium Solvers Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A custom and lightweight neural solver for deep equilibrium models significantly improves their efficiency with minimal training.	Shaojie Bai; Vladlen Koltun; J Zico Kolter;
561	Neural Energy Minimization for Molecular Conformation Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a neural energy minimization formulation that casts the prediction problem into an unrolled optimization process, where a neural network is parametrized to learn the gradient fields of a conformational energy landscape.	Jiaqi Guan; Wesley Wei Qian; qiang liu; Wei-Ying Ma; Jianzhu Ma; Jian Peng;
562	Neural Graphical Modelling in Continuous-time: Consistency Guarantees and Algorithms Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present algorithms and consistency guarantees for graphical modelling in dynamical systems.	Alexis Bellot; Kim Branson; Mihaela van der Schaar;
563	Neural Link Prediction with Walk Pooling Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a link prediction algorithm based on a new pooling scheme called WalkPool.	Liming Pan; Cheng Shi; Ivan Dokmanic;
564	Neural Markov Controlled SDE: Stochastic Optimization for Continuous-Time Data Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel probabilistic framework for modelling stochastic dynamics with the rigorous use of optimal control theory.	Sung Woo Park; Kyungjae Lee; Junseok Kwon;
565	Neural Methods for Logical Reasoning Over Knowledge Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Neural Network models for answering multi-hop queris on Knowledge Graphs We introduce a set of models that use Neural Networks to create one-point vector embeddings to answer the queries.	Alfonso Amayuelas; Shuai Zhang; Xi Susie Rao; Ce Zhang;
566	Neural Models for Output-Space Invariance in Combinatorial Problems Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We design GNN based neural models to achieve output space invariance in combinatorial problems, e.g. solving 16 x 16 sudoku after training on 9 x 9 sudoku	Yatin Nandwani; Vidit Jain; Mausam .; Parag Singla;
567	Neural Network Approximation Based on Hausdorff Distance of Zonotopes Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Based on this result, we propose geometrical neural network compression methods that employ the K-means algorithm.	Panagiotis Misiakos; Georgios Smyrnis; George Retsinas; Petros Maragos;
568	Neural Networks As Kernel Learners: The Silent Alignment Effect Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Neural networks with small initialization, trained in the rich feature learning regime, can learn kernel regression solutions for a data adaptive kernel.	Alexander Atanasov; Blake Bordelon; Cengiz Pehlevan;
569	Neural Parameter Allocation Search Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: An efficient approach for searching for the optimal allocation of parameters to layers across any neural network	Bryan A. Plummer; Nikoli Dryden; Julius Frost; Torsten Hoefler; Kate Saenko;
570	Neural Processes with Stochastic Attention: Paying More Attention to The Context Dataset Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper extends the attentive neural process (ANP), replacing the deterministic weights in the cross-attention module of ANP with latent weights.	Mingyu Kim; Kyeong Ryeol Go; Se-Young Yun;
571	Neural Program Synthesis with Query Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a query-based framework for the interactive program synthesis.	Di Huang; Rui Zhang; Xing Hu; Xishan Zhang; Pengwei Jin; Nan Li; Zidong Du; Qi Guo; Yunji Chen;
572	Neural Relational Inference with Node-Specific Information Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We use variational inference to uncover relations among agents in a multi-agent system, given that the agents can have access to some private information	Ershad Banijamali;
573	Neural Solvers for Fast and Accurate Numerical Optimal Control Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a hypersolvers approach for numerical optimal control which shows consistent Pareto improvements in solution accuracy and control performance.	Federico Berto; Stefano Massaroli; Michael Poli; Jinkyoo Park;
574	Neural Spectral Marked Point Processes Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we introduce a novel and general neural network-based non-stationary influence kernel with high expressiveness for handling complex discrete events data while providing theoretical performance guarantees.	Shixiang Zhu; Haoyun Wang; Zheng Dong; Xiuyuan Cheng; Yao Xie;
575	Neural Stochastic Dual Dynamic Programming Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We proposed neural-SDDP pushing the frontier of multi-stage stochastic optimization solver towards practical problem size.	Hanjun Dai; Yuan Xue; Zia Syed; Dale Schuurmans; Bo Dai;
576	New Insights on Reducing Abrupt Representation Change in Online Continual Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study how representations shift at task boundaries in the single-head online continual learning setting, leading to a simple high performance method	Lucas Caccia; Rahaf Aljundi; Nader Asadi; Tinne Tuytelaars; Joelle Pineau; Eugene Belilovsky;
577	Node Feature Extraction By Self-Supervised Multi-scale Neighborhood Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We design a self-supervised learning method for extracting node representations from raw data.	Eli Chien; Wei-Cheng Chang; Cho-Jui Hsieh; Hsiang-Fu Yu; Jiong Zhang; Olgica Milenkovic; Inderjit S Dhillon;
578	NODE-GAM: Neural Generalized Additive Model for Interpretable Deep Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We develop a deep-learning version of Generalized Additive Model (GAM) and GA2M that is accurate, scalable and interpretable.	Chun-Hao Chang; Rich Caruana; Anna Goldenberg;
579	NodePiece: Compositional and Parameter-Efficient Representations of Large Knowledge Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Node hashing in graphs for 10-100x embedding size reduction without significant performance losses on many tasks and inductive out of the box.	Mikhail Galkin; Etienne Denis; Jiapeng Wu; William L. Hamilton;
580	Noisy Feature Mixup Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose and study Noisy Feature Mixup, a simple yet effective data augmentation method that leads to improved model robustness when compared to training with manifold mixup or noise injection alone.	Soon Hoe Lim; N. Benjamin Erichson; Francisco Utrera; Winnie Xu; Michael W. Mahoney;
581	Non-Autoregressive Models Are Better Multilingual Translators Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose switch-GLAT, a non-autoregressive multilingual machine translation model with a code-switch decoder.	Zhenqiao Song; Hao Zhou; Lihua Qian; Jingjing Xu; Shanbo Cheng; Mingxuan Wang; Lei Li;
582	Non-Linear Operator Approximations for Initial Value Problems Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A Pad? approximation based exponential operator module is proposed for working with the Initial Value Problems. The compactness of model yields data-efficiency and better performance which is demonstrated on scarce real-world dataset.	Gaurav Gupta; Xiongye Xiao; Radu Balan; Paul Bogdan;
583	Non-Parallel Text Style Transfer with Self-Parallel Supervision Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new text style transfer model for non-parallel corpus with supervision from intrinsic parallelism.	Ruibo Liu; Chongyang Gao; Chenyan Jia; Guangxuan Xu; Soroush Vosoughi;
584	No One Representation to Rule Them All: Overlapping Features of Training Methods Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study the effect of training methodology on prediction diversity and show that diverging training setups produce diverse features, uncorrelated errors, and more efficient ensembles.	Raphael Gontijo-Lopes; Yann Dauphin; Ekin Dogus Cubuk;
585	No Parameters Left Behind: Sensitivity Guided Adaptive Learning Rate for Training Large Transformer Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel adaptive learning rate training strategy for large Transformer models that encourages all parameters to be trained sufficiently.	Chen Liang; Haoming Jiang; Simiao Zuo; Pengcheng He; Xiaodong Liu; Jianfeng Gao; Weizhu Chen; Tuo Zhao;
586	OBJECT DYNAMICS DISTILLATION FOR SCENE DECOMPOSITION AND REPRESENTATION Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we work on object dynamics and propose Object Dynamics Distillation Network (ODDN), a framework that distillates explicit object dynamics (e.g., velocity) from sequential static representations.	Qu Tang; Xiangyu Zhu; Zhen Lei; Zhaoxiang Zhang;
587	Object Pursuit: Building A Space of Objects Via Discriminative Weight Generation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel framework named object pursuit that can continuously learn object-centric representations using training data collected from interactions with individual objects.	Chuanyu Pan; Yanchao Yang; Kaichun Mo; Yueqi Duan; Leonidas Guibas;
588	Objects in Semantic Topology Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we provide a unified perspective: Semantic Topology.	Shuo Yang; Peize Sun; Yi Jiang; Xiaobo Xia; Ruiheng Zhang; Zehuan Yuan; Changhu Wang; Ping Luo; Min Xu;
589	Offline Neural Contextual Bandits: Pessimism, Optimization and Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Provably efficient offline contextual bandits with neural network function approximation	Thanh Nguyen-Tang; Sunil Gupta; A. Tuan Nguyen; Svetha Venkatesh;
590	Offline Reinforcement Learning for Large Scale Language Action Spaces Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we introduce GPT-Critic, an offline RL method for task-oriented dialogue.	Youngsoo Jang; Jongmin Lee; Kee-Eung Kim;
591	Offline Reinforcement Learning with In-sample Q-Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Offline RL method with only dataset actions.	Ilya Kostrikov; Ashvin Nair; Sergey Levine;
592	Offline Reinforcement Learning with Value-based Episodic Memory Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new offline RL method which uses expectile value learning and memory-based planning.	Xiaoteng Ma; Yiqin Yang; Hao Hu; Jun Yang; Chongjie Zhang; Qianchuan Zhao; Bin Liang; Qihan Liu;
593	Omni-Dimensional Dynamic Convolution Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper presents Omni-dimensional Dynamic Convolution (ODConv) to advance the research in dynamic convolution.	Chao Li; Aojun Zhou; Anbang Yao;
594	Omni-Scale CNNs: A Simple and Effective Kernel Size Configuration for Time Series Classification Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To extract features from time series data in proper time scales, many complicated scales searching or weighting methods have been proposed, but we will show that this could have been done via a very simple structure.	Wensi Tang; Guodong Long; Lu Liu; Tianyi Zhou; Michael Blumenstein; Jing Jiang;
595	On Bridging Generic and Personalized Federated Learning for Image Classification Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Concretely, we propose a novel federated learning framework that explicitly decouples a model’s dual duties with two prediction tasks.	Hong-You Chen; Wei-Lun Chao;
596	On Covariate Shift of Latent Confounders in Imitation and Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We use expert data with unobserved confounders for both imitation and reinforcement learning. Such hidden confounding is prone to a shifted distribution, which may severely hurt performance unless accounted for.	Guy Tennenholtz; Assaf Hallak; Gal Dalal; Shie Mannor; Gal Chechik; Uri Shalit;
597	On Distributed Adaptive Optimization with Gradient Compression Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study COMP-AMS, a distributed optimization framework based on gradient averaging and adaptive AMSGrad algorithm.	Xiaoyun Li; Belhal Karimi; Ping Li;
598	One After Another: Learning Incremental Skills for A Changing World Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We discover skills incrementally, without supervision, and it lets us learn skills in both static environments and environments with changing dynamics.	Nur Muhammad Mahi Shafiullah; Lerrel Pinto;
599	On Evaluation Metrics for Graph Generative Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we mitigate these issues by finding \emph{a scalar, domain-agnostic, and scalable metric} for evaluating and ranking GGMs.	Rylee Thompson; Boris Knyazev; Elahe Ghalebi; Jungtaek Kim; Graham W. Taylor;
600	On Feature Learning in Shallow and Multi-layer Neural Networks with Global Convergence Guarantees Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Gradient flow can induce feature learning in shallow and multi-layer neural networks while admitting non-asymptotic guarantees of global convergence.	Zhengdao Chen; Eric Vanden-Eijnden; Joan Bruna;
601	On Improving Adversarial Transferability of Vision Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Novel approach to improve transferability of adversarial perturbations found in vision transformers via self-ensemble and token refinement.	Muzammal Naseer; Kanchana Ranasinghe; Salman Khan; Fahad Khan; Fatih Porikli;
602	On Incorporating Inductive Biases Into VAEs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A flexible and effective framework for adding inductive biases to VAEs that corrects the pathologies of previous approaches and leads to improved representations and generative models.	Ning Miao; Emile Mathieu; Siddharth N; Yee Whye Teh; Tom Rainforth;
603	Online Ad Hoc Teamwork Under Partial Observability Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To overcome partial observability, we introduce an information-based regularizer to derive \emph{proxy} representations of the learned variables from local observations.	Pengjie Gu; Mengchen Zhao; Jianye Hao; Bo An;
604	Online Adversarial Attacks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We consider a new adversarial attack setting in which the data arrives as a stream and an adversary must pick top-k items to craft blackbox transfer attacks against an unkown target model.	Andjela Mladenovic; Joey Bose; Hugo berard; William L. Hamilton; Simon Lacoste-Julien; Pascal Vincent; Gauthier Gidel;
605	Online Continual Learning on Class Incremental Blurry Task Configuration with Anytime Inference Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: a novel continual learning set-up that is online and task-free, has new class distributions and focuses on any-time inference	Hyunseo Koh; Dahyun Kim; Jung-Woo Ha; Jonghyun Choi;
606	Online Coreset Selection for Rehearsal-based Continual Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Online Coreset Selection (OCS), a simple yet effective method that selects the most representative and informative coreset at each iteration and trains them in an online manner.	Jaehong Yoon; Divyam Madaan; Eunho Yang; Sung Ju Hwang;
607	Online Facility Location with Predictions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We give a nearly optimal robust algorithm for online facility location with predictions.	Shaofeng H.-C. Jiang; Erzhi Liu; You Lyu; Zhihao Gavin Tang; Yubo Zhang;
608	Online Hyperparameter Meta-Learning with Hypergradient Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a gradient-based hyperparameter optimization method based on the idea of knowledge distillation, which is fully online and applicable to high-dimensional hyperparameters.	Hae Beom Lee; Hayeon Lee; JaeWoong Shin; Eunho Yang; Timothy Hospedales; Sung Ju Hwang;
609	Online Target Q-learning with Reverse Experience Replay: Efficiently Finding The Optimal Policy for Linear MDPs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The goal of this work is to bridge the gap between practical success of Q-learning and the relatively pessimistic theoretical results.	Naman Agarwal; Syomantak Chaudhuri; Prateek Jain; Dheeraj Mysore Nagaraj; Praneeth Netrapalli;
610	On Lottery Tickets and Minimal Task Representations in Deep Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We investigate the mechanisms underlying the lottery ticket effect in Deep RL and show that the derived mask extracts minimal task representations.	Marc Vischer; Robert Tjarko Lange; Henning Sprekeler;
611	On Non-Random Missing Labels in Semi-Supervised Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We presented a principled class-aware doubly robust solution to handle the non-random missing labels in semi-supervised learning.	Xinting Hu; Yulei Niu; Chunyan Miao; Xian-Sheng Hua; Hanwang Zhang;
612	On-Policy Model Errors in Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We combine real-world data and a learned model for data-efficient reinforcement learning with reduced model-bias.	Lukas Froehlich; Maksym Lefarov; Melanie Zeilinger; Felix Berkenkamp;
613	On Predicting Generalization Using GANs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The current paper investigates a simple idea: can test error be predicted using {\em synthetic data,} produced using a Generative Adversarial Network (GAN) that was trained on the same training dataset?	Yi Zhang; Arushi Gupta; Nikunj Saunshi; Sanjeev Arora;
614	On Redundancy and Diversity in Cell-based Neural Architecture Search Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We analyse and explore the redundancies and diversity of popular cell-based search spaces in NAS.	Xingchen Wan; Binxin Ru; Pedro M Esperan�a; Zhenguo Li;
615	On Robust Prefix-Tuning for Text Classification Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a robust prefix-tuning framework that improves robustness of prefix-tuning against different types of attacks while preserving its efficiency and modularity with interpretation from the perspective of optimal control.	Zonghan Yang; Yang Liu;
616	On The Approximation Properties of Recurrent Encoder-decoder Architectures Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Approximation properties of recurrent encoder-decoder architectures are given, where the formed temporal product structure further characterises temporal relationships able to be efficiently learned.	Zhong Li; Haotian Jiang; Qianxiao Li;
617	On The Benefits of Maximum Likelihood Estimation for Regression and Forecasting Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: MLE + Post-Hoc Inference can be competitive with any estimator under some general assumptions	Pranjal Awasthi; Abhimanyu Das; Rajat Sen; Ananda Theertha Suresh;
618	On The Certified Robustness for Ensemble Models and Beyond Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Inspired by theoretical analysis, we propose Diversity Regularized Training to enhance the certified robustness of ensemble models and DRT significantly outperforms existing methods.	Zhuolin Yang; Linyi Li; Xiaojun Xu; Bhavya Kailkhura; Tao Xie; Bo Li;
619	On The Connection Between Local Attention and Dynamic Depth-wise Convolution Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study the connection between local attention and dynamic depth-wise convolution in terms of sparse connectivity, weight sharing, and dynamic weight	Qi Han; Zejia Fan; Qi Dai; Lei Sun; Ming-Ming Cheng; Jiaying Liu; Jingdong Wang;
620	On The Convergence of Certified Robust Training with Interval Bound Propagation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present the first theoretical analysis on the convergence of certified robust training with interval bound propagation.	Yihan Wang; Zhouxing Shi; Quanquan Gu; Cho-Jui Hsieh;
621	On The Convergence of MSGD and AdaGrad for Stochastic Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A theoretical paper focusing on the investigation for the convergence of mSGD and AdaGrad optimization algorithms.	ruinan Jin; Yu Xing; Xingkang He;
622	On The Convergence of The Monte Carlo Exploring Starts Algorithm for Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We prove that the Monte Carlo Exploring Starts algorithm converges for optimal policy feed-forward MDPs.	Che Wang; Shuhan Yuan; Kai Shao; Keith W. Ross;
623	On The Existence of Universal Lottery Tickets Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We formalize a notion of strong universal lottery tickets and prove their existence as subnetworks of randomly initialized neural networks.	Rebekka Burkholz; Nilanjana Laha; Rajarshi Mukherjee; Alkis Gotovos;
624	On The Importance of Difficulty Calibration in Membership Inference Attacks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Membership inference attacks can greatly benefit from a technique called difficulty calibration, significantly improving their reliability.	Lauren Watson; Chuan Guo; Graham Cormode; Alexandre Sablayrolles;
625	On The Importance of Firth Bias Reduction in Few-Shot Classification Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we demonstrate the effectiveness of Firth bias reduction in few-shot classification.	Saba Ghaffari; Ehsan Saleh; David Forsyth; Yu-Xiong Wang;
626	On The Learning of Quasimetrics Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We theoretically analyze various algorithms on learning quasimetrics (asymmetrical metrics), and propose an embedding-based method with strong guarantees. Experiments on graph learning and Q-learning show its effectiveness over common baselines.	Tongzhou Wang; Phillip Isola;
627	On The Limitations of Multimodal VAEs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We prove that the sub-sampling of modalities enforces an undesirable upper bound on the multimodal ELBO and thereby limits the generative quality of the respective models.	Imant Daunhawer; Thomas M. Sutter; Kieran Chin-Cheong; Emanuele Palumbo; Julia E Vogt;
628	On The Optimal Memorization Power of ReLU Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that ReLU neural networks can memorize N samples using \sqrt{N} parameters, and prove that up to logarithmic terms this is the optimal solution.	Gal Vardi; Gilad Yehudai; Ohad Shamir;
629	On The Pitfalls of Analyzing Individual Neurons in Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We analyze and compare methods to rank neurons in hidden representations according to their relevance to morphologic attributes, and show some of their weaknesses.	Omer Antverg; Yonatan Belinkov;
630	On The Pitfalls of Heteroscedastic Uncertainty Estimation with Probabilistic Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We analyse problems with the training objective of probabilistic neural networks and propose a fix in the form of a new loss function.	Maximilian Seitzer; Arash Tavakoli; Dimitrije Antic; Georg Martius;
631	On The Relation Between Statistical Learning and Perceptual Distances Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we aim to unravel the non-trivial relationships between the probability distribution of the data, perceptual distances, and unsupervised machine learning.	Alexander Hepburn; Valero Laparra; Raul Santos-Rodriguez; Johannes Ball�; Jesus Malo;
632	On The Uncomputability of Partition Functions in Energy-Based Sequence Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: EBMs over sequences have several theoretical limitations as learnable probabilistic sequence models.	Chu-Cheng Lin; Arya D. McCarthy;
633	OntoProtein: Protein Pretraining With Gene Ontology Embedding Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A general framework to integrate knowledge graph (gene ontology) into protein pre-training.	Ningyu Zhang; Zhen Bi; Xiaozhuan Liang; Siyuan Cheng; Haosen Hong; Shumin Deng; Qiang Zhang; Jiazhang Lian; Huajun Chen;
634	Open-vocabulary Object Detection Via Vision and Language Knowledge Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose using knowledge distillation to train an object detector that can detect objects with arbitrary text inputs, outperforming its supervised counterparts on rare categories.	Xiuye Gu; Tsung-Yi Lin; Weicheng Kuo; Yin Cui;
635	Open-World Semi-Supervised Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a pipeline that recognizes previously seen classes and discovers novel, never-before-seen classes at the same time.	Kaidi Cao; Maria Brbic; Jure Leskovec;
636	Optimal ANN-SNN Conversion for High-accuracy and Ultra-low-latency Spiking Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: An ANN-SNN conversion method enables high-accuracy and ultra-low-latency deep SNNs.	Tong Bu; Wei Fang; Jianhao Ding; PENGLIN DAI; Zhaofei Yu; Tiejun Huang;
637	Optimal Representations for Covariate Shift Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We give a simple variational objective whose optima are exactly the set of representations that are robust under covariate shift	Yangjun Ruan; Yann Dubois; Chris J. Maddison;
638	Optimal Transport for Causal Discovery Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we provide a novel dynamical-system view of FCMs and propose a new framework for identifying causal direction in the bivariate case.	Ruibo Tu; Kun Zhang; Hedvig Kjellstrom; Cheng Zhang;
639	Optimal Transport for Long-Tailed Recognition with Learnable Cost Matrix Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A new paradigm for post-hoc correction based on optimal transport to cope with long-tailed recognition.	Hanyu Peng; Mingming Sun; Ping Li;
640	Optimization and Adaptive Generalization of Three Layer Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Algorithmically obtaining noise-robust and adaptive generalization bounds for a three layer network model by going beyond the linear approximation of the network	Khashayar Gatmiry; Stefanie Jegelka; Jonathan Kelner;
641	Optimization Inspired Multi-Branch Equilibrium Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A Multi-Branch DEQ model and its training strategy proposed by minimizing an objective function designed as our purpose.	Mingjie Li; Yisen Wang; Xingyu Xie; Zhouchen Lin;
642	Optimizer Amalgamation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study a new problem named Optimizer Amalgamation: combining a pool of teacher optimizers into a single student optimizer which can have stronger problem-specific performance.	Tianshu Huang; Tianlong Chen; Sijia Liu; Shiyu Chang; Lisa Amini; Zhangyang Wang;
643	Optimizing Few-Step Diffusion Samplers By Gradient Descent Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a method to discover fast, high-fidelity samplers for diffusion probabilistic models.	Daniel Watson; William Chan; Jonathan Ho; Mohammad Norouzi;
644	Optimizing Neural Networks with Gradient Lexicase Selection Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Gradient Lexicase Selection, an evolutionary optimization method that improves the generalization of deep neural networks.	Li Ding; Lee Spector;
645	Orchestrated Value Mapping for Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a general convergent class of RL algorithms based on combining arbitrary value mappings and reward decomposition.	Mehdi Fatemi; Arash Tavakoli;
646	PAC-Bayes Information Bottleneck Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel PAC-Bayes bound guided information bottleneck for understanding and enhancing deep representation learning.	Zifeng Wang; Shao-Lun Huang; Ercan Engin Kuruoglu; Jimeng Sun; Xi Chen; Yefeng Zheng;
647	PAC Prediction Sets Under Covariate Shift Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel algorithm that constructs a prediction set with probably approximated correct (PAC) guarantee under covariate shift, while minimizing the expected prediction set size.	Sangdon Park; Edgar Dobriban; Insup Lee; Osbert Bastani;
648	P-Adapters: Robustly Extracting Factual Information from Language Models with Diverse Prompts Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We make prompting large language models for factual information more robust by introducing P-Adapter models.	Benjamin Newman; Prafulla Kumar Choubey; Nazneen Rajani;
649	Parallel Training of GRU Networks with A Multi-Grid Solver for Long Sequences Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper presents a novel parallel-in-time training scheme for GRU networks based on a MGRIT solver.	Euhyun Moon; Eric C Cyr;
650	Pareto Policy Adaptation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a policy gradient method for multi-objective reinforcement learning under unknown, linear preferences.	Panagiotis Kyriakis; Jyotirmoy Deshmukh; Paul Bogdan;
651	Pareto Policy Pool for Model-based Offline Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a model-based offline RL method that builds a diverse set of optimal policies on Pareto front providing different levels of model return-uncertainty trade-off and it significantly outperforms single-policy methods.	Yijun Yang; Jing Jiang; Tianyi Zhou; Jie Ma; Yuhui Shi;
652	Partial Wasserstein Adversarial Network for Non-rigid Point Set Registration Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a method for large scale partial distribution matching problem, and apply it to non-rigid point set registration task.	Ziming Wang; Nan Xue; Ling Lei; Gui-Song Xia;
653	Patch-Fool: Are Vision Transformers Always Robust Against Adversarial Perturbations? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the Patch-Fool attack to unveil a vulnerability perspective of ViTs.	Yonggan Fu; Shunyao Zhang; Shang Wu; Cheng Wan; Yingyan Lin;
654	Path Auxiliary Proposal for MCMC in Discrete Space Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we present a path auxiliary algorithm that uses a composition of local moves to efficiently explore large neighborhoods.	Haoran Sun; Hanjun Dai; Wei Xia; Arun Ramamurthy;
655	Path Integral Sampler: A Stochastic Control Approach For Sampling Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present Path Integral Sampler~(PIS), an efficient algorithm to draw samples from unnormalized probability density functions.	Qinsheng Zhang; Yongxin Chen;
656	PEARL: Data Synthesis Via Private Embeddings and Adversarial Reconstruction Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new framework of synthesizing data using deep generative models in a differentially private manner.	Seng Pei Liew; Tsubasa Takahashi; Michihiko Ueno;
657	Perceiver IO: A General Architecture for Structured Inputs & Outputs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Perceiver IO, a general-purpose architecture that handles data from arbitrary settings while scaling linearly with the size of inputs and outputs.	Andrew Jaegle; Sebastian Borgeaud; Jean-Baptiste Alayrac; Carl Doersch; Catalin Ionescu; David Ding; Skanda Koppula; Daniel Zoran; Andrew Brock; Evan Shelhamer; Olivier J Henaff; Matthew Botvinick; Andrew Zisserman; Oriol Vinyals; Joao Carreira;
658	PER-ETD: A Polynomially Efficient Emphatic Temporal Difference Learning Method Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new off-policy evaluation algorithm called PER-ETD (i.e., PEriodically Restarted Emphatic TD), which improves upon its precursor ETD with reduced variance and polynomial sample efficiency.	Ziwei Guan; Tengyu Xu; Yingbin Liang;
659	Permutation-Based SGD: Is Random Optimal? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show that the question of whether random permutations are optimal for permutation-based SGD is nuanced, and depends on the family of functions one is trying to optimize.	Shashank Rajput; Kangwook Lee; Dimitris Papailiopoulos;
660	Permutation Compressors for Provably Faster Distributed Nonconvex Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we present the novel compression scheme for distributed non-convex optimization.	Rafal Szlendak; Alexander Tyurin; Peter Richt�rik;
661	Pessimistic Bootstrapping for Uncertainty-Driven Offline Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose pessimistic bootstrapping as a purely uncertainty-driven algorithm for offline Reinforcement Learning.	Chenjia Bai; Lingxiao Wang; Zhuoran Yang; Zhi-Hong Deng; Animesh Garg; Peng Liu; Zhaoran Wang;
662	Pessimistic Model-based Offline Reinforcement Learning Under Partial Coverage Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study model-based offline Reinforcement Learning with general function approximation without a full coverage assumption on the offline data distribution.	Masatoshi Uehara; Wen Sun;
663	PF-GNN: Differentiable Particle Filtering Based Approximation of Universal Graph Representations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Increasing the expressive power of Graph Neural Networks by using techniques from exact isomorphism solvers with a particle filtering approach.	Mohammed Haroon Dupty; Yanfei Dong; Wee Sun Lee;
664	Phase Collapse in Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: The classification accuracy of CNNs mostly relies on the mechanism of phase collapses to eliminate spatial variability and linearly separate class means.	Florentin Guth; John Zarka; St�phane Mallat;
665	Phenomenology of Double Descent in Finite-Width Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide a theoretical analysis of double descent that applies for finite-width neural networks and delivers insights into the properties of neural networks (and its Hessian spectrum) near the interpolation threshold.	Sidak Pal Singh; Aurelien Lucchi; Thomas Hofmann; Bernhard Sch�lkopf;
666	PI3NN: Out-of-distribution-aware Prediction Intervals from Three Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel prediction interval (PI) method for uncertainty quantification, which addresses three major issues with the state-of-the-art PI methods.	Siyan Liu; Pei Zhang; Dan Lu; Guannan Zhang;
667	$\pi$BO: Augmenting Acquisition Functions with User Beliefs for Bayesian Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We extend the Bayesian Optimization framework by allowing for arbitrary user priors over promising regions of the search space, to guide the search towards said regions.	Carl Hvarfner; Danny Stoll; Artur Souza; Luigi Nardi; Marius Lindauer; Frank Hutter;
668	Pix2seq: A Language Modeling Framework for Object Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We demonstrated that object detection can be tackled by simply training a language model conditioned on pixel inputs.	Ting Chen; Saurabh Saxena; Lala Li; David J. Fleet; Geoffrey Hinton;
669	Pixelated Butterfly: Simple and Efficient Sparse Training for Neural Network Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a simple sparse training method, which can speed up model training in wall-clock time with no drop in accuracy.	Beidi Chen; Tri Dao; Kaizhao Liang; Jiaming Yang; Zhao Song; Atri Rudra; Christopher Re;
670	Planning in Stochastic Environments with A Learned Model Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Specifically, we introduce a new algorithm, Stochastic MuZero, that learns a stochastic model incorporating afterstates, and uses this model to perform a stochastic tree search.	Ioannis Antonoglou; Julian Schrittwieser; Sherjil Ozair; Thomas K Hubert; David Silver;
671	Plant ‘n’ Seek: Can You Find The Winning Ticket? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We derive a framework to plant ground truth lottery tickets in randomly initialized deep neural networks.	Jonas Fischer; Rebekka Burkholz;
672	POETREE: Interpretable Policy Learning with Adaptive Decision Trees Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Policy Extraction through decision Trees (POETREE) is a novel framework for interpretable policy learning, compatible with fully-offline and partially-observable clinical decision environments.	Aliz�e Pace; Alex Chan; Mihaela van der Schaar;
673	Poisoning and Backdooring Contrastive Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We argue poisoning and backdooring attacks are a serious threat to multimodal contrastive classifiers, because they are explicitly designed to be trained on uncurated datasets from the Internet.	Nicholas Carlini; Andreas Terzis;
674	Policy Gradients Incorporating The Future Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Namely, we propose to allow an agent, during its training on past experience, to observe what \emph{actually} happened in the future at that time, while enforcing an information bottleneck to avoid the agent overly relying on this privileged information.	David Venuto; Elaine Lau; Doina Precup; Ofir Nachum;
675	Policy Improvement By Planning with Gumbel Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We redesign AlphaZero to keep improving even when training with a small number of simulations.	Ivo Danihelka; Arthur Guez; Julian Schrittwieser; David Silver;
676	Policy Smoothing for Provably Robust Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A provable adversarial robustness technique for reinforcement learning.	Aounon Kumar; Alexander Levine; Soheil Feizi;
677	PolyLoss: A Polynomial Expansion Perspective of Classification Loss Functions Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In the PolyLoss framework, we propose a simple and effective Poly-1 formulation which outperforms the cross-entropy loss and focal loss on various of tasks.	Zhaoqi Leng; Mingxing Tan; Chenxi Liu; Ekin Dogus Cubuk; Jay Shi; Shuyang Cheng; Dragomir Anguelov;
678	PoNet: Pooling Network for Efficient Token Mixing in Long Sequences Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel Pooling Network for token mixing with linear complexity, achieve competitive performance on the Long Range Arena benchmark, and 95.7% of the accuracy of BERT on the GLUE demonstrating its transferability.	Chao-Hong Tan; Qian Chen; Wen Wang; Qinglin Zhang; Siqi Zheng; Zhen-Hua Ling;
679	Possibility Before Utility: Learning And Using Hierarchical Affordances Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a method that achieves superior performance in complex hierarchical tasks by utilizing a notion of subtask dependency grounded in the present state.	Robby Costales; Shariq Iqbal; Fei Sha;
680	Post-Training Detection of Backdoor Attacks for Two-Class and Multi-Attack Scenarios Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We proposed a detection framework against backdoor attacks for two-class and multi-attack scenarios, without access to the classifier’s training set or any supervision from clean classifiers trained for the same domain.	Zhen Xiang; David Miller; George Kesidis;
681	Practical Conditional Neural Process Via Tractable Dependent Predictions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we present a new class of Neural Process models that produce correlated predictions and support exact maximum likelihood training that is simple and scalable.	Stratis Markou; James Requeima; Wessel Bruinsma; Anna Vaughan; Richard E Turner;
682	Practical Integration Via Separable Bijective Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We explore a method that enables learning over hypervolumes within the data space.	Christopher M Bender; Patrick Emmanuel; Michael K. Reiter; Junier Oliva;
683	Predicting Physics in Mesh-reduced Space with Temporal Attention Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We use a GNN to locally summarize features and create coarsened, compact mesh representation of the system state, onto which we apply a transformer-style temporal attention module for physics prediction.	XU HAN; Han Gao; Tobias Pfaff; Jian-Xun Wang; Liping Liu;
684	Predictive Modeling in The Presence of Nuisance-Induced Spurious Correlations Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper build models robust to nuisance-induced spurious correlations by constructing a representation that distills out the influence of the nuisance variables, while also maximizing its information with the label.	Aahlad Manas Puli; Lily H Zhang; Eric Karl Oermann; Rajesh Ranganath;
685	Preference Conditioned Neural Multi-objective Combinatorial Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a learning-based method to approximate the whole Pareto set for multi-objective combinatorial optimization problems with a single model.	Xi Lin; Zhiyuan Yang; Qingfu Zhang;
686	Pre-training Molecular Graph Representation with 3D Geometry Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We proposed a new SSL framework to make 3D geomety information helpful for 2D representation, in terms of the downstream tasks with 2D info only.	Shengchao Liu; Hanchen Wang; Weiyang Liu; Joan Lasenby; Hongyu Guo; Jian Tang;
687	Pretraining Text Encoders with Adversarial Mixture of Training Signal Generators Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present AMOS, a new method that pretrains text encoders with an Adversarial learning curriculum via a Mixture Of Signals from multiple auxiliary generators.	Yu Meng; Chenyan Xiong; Payal Bajaj; saurabh tiwary; Paul N. Bennett; Jiawei Han; Xia Song;
688	PriorGrad: Improving Conditional Denoising Diffusion Models with Data-Dependent Adaptive Prior Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We improve the efficiency of diffusion-based conditional generative models for audio by using data-dependent non-standard Gaussian as a prior.	Sang-gil Lee; Heeseung Kim; Chaehun Shin; Xu Tan; Chang Liu; Qi Meng; Tao Qin; Wei Chen; Sungroh Yoon; Tie-Yan Liu;
689	Privacy Implications of Shuffling Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: a novel formalization of the privacy offered by shuffling	Casey Meehan; Amrita Roy Chowdhury; Kamalika Chaudhuri; Somesh Jha;
690	Probabilistic Implicit Scene Completion Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a scalable generative model for multi-modal completion of 3D scenes in implicit representation.	Dongsu Zhang; Changwoon Choi; Inbum Park; Young Min Kim;
691	Procedural Generalization By Planning with Self-supervised World Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study generalization in model-based agents and find that they excel at procedural generalization, with planning, self-supervision and data-diversity combining to yield SoTA results on Procgen; however, task generalization is more challenging.	Ankesh Anand; Jacob C Walker; Yazhe Li; Eszter V�rtes; Julian Schrittwieser; Sherjil Ozair; Theophane Weber; Jessica B Hamrick;
692	Programmatic Reinforcement Learning Without Oracles Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a differentiable program architecture search framework to synthesize interpretable, generalizable, and compositional programs for controlling reinforcement learning applications.	Wenjie Qiu; He Zhu;
693	Progressive Distillation for Fast Sampling of Diffusion Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Diffusion models now need just 4 sampling steps to produce high quality samples.	Tim Salimans; Jonathan Ho;
694	Promoting Saliency From Depth: Deep Unsupervised RGB-D Saliency Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the first deep unsupervised RGB-D saliency detection method, which achieves appealing performance and does not require any human efforts compared to fully-supervised learning.	Wei Ji; Jingjing Li; Qi Bi; chuan guo; Jie Liu; Li Cheng;
695	Proof Artifact Co-Training for Theorem Proving with Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Co-training on low-level proof tasks in a Transformer theorem prover improves success from 32% to 48% on Lean theorems.	Jesse Michael Han; Jason Rute; Yuhuai Wu; Edward Ayers; Stanislas Polu;
696	Properties from Mechanisms: An Equivariance Perspective on Identifiable Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Representation learning is identifiable up to any equivariances of the (known) mechanisms that govern an environment’s evolution.	Kartik Ahuja; Jason Hartford; Yoshua Bengio;
697	Prospect Pruning: Finding Trainable Weights at Initialization Using Meta-Gradients Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We use meta-gradients to prune neural networks at initialization based on trainability of weights instead of their impact on the loss at a single step.	Milad Alizadeh; Shyam A. Tailor; Luisa M Zintgraf; Joost van Amersfoort; Sebastian Farquhar; Nicholas Donald Lane; Yarin Gal;
698	Prototypical Contrastive Predictive Coding Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose prototypical contrastive predictive coding for efficient distillation of representational knowledge of one network into other network.	Kyungmin Lee;
699	Provable Adaptation Across Multiway Domains Via Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a model which consists of a domain-invariant latent representation layer and a domain-specific linear prediction layer with a low-rank tensor structure.	Zhili Feng; Shaobo Han; Simon Shaolei Du;
700	Provable Learning-based Algorithm For Sparse Recovery Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we propose PLISA (Provable Learning-based Iterative Sparse recovery Algorithm) to learn algorithms automatically from data.	Xinshi Chen; Haoran Sun; Le Song;
701	Provably Convergent Quasistatic Dynamics for Mean-field Two-player Zero-sum Games Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a quasistatic Wasserstein flow for finding mixed Nash equilibriums, and prove its convergence.	Chao Ma; Lexing Ying;
702	Provably Robust Adversarial Examples Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce the concept of provably robust adversarial examples $-$ connected input regions constructed from standard adversarial examples and guaranteed to be provably robust to a set of perturbations.	Dimitar Iliev Dimitrov; Gagandeep Singh; Timon Gehr; Martin Vechev;
703	Proving The Lottery Ticket Hypothesis for Convolutional Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We prove the lottery ticket hypothesis for convolutional neural networks	Arthur da Cunha; Emanuele Natale; Laurent Viennot;
704	PSA-GAN: Progressive Self Attention GANs for Synthetic Time Series Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper we present PSA-GAN, a generative adversarial network (GAN) that generates long time series samples of high quality using progressive growing of GANs and self-attention.	Paul Jeha; Michael Bohlke-Schneider; Pedro Mercado; Shubham Kapoor; Rajbir Singh Nirwan; Valentin Flunkert; Jan Gasthaus; Tim Januschowski;
705	Pseudo-Labeled Auto-Curriculum Learning for Semi-Supervised Keypoint Localization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The first work that explores automatic curriculum learning for semi-supervised keypoint localization	Can Wang; Sheng Jin; Yingda Guan; Wentao Liu; Chen Qian; Ping Luo; Wanli Ouyang;
706	Pseudo Numerical Methods for Diffusion Models on Manifolds Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose PNDMs, a new kind of numerical method, to accelerate diffusion models on manifolds.	Luping Liu; Yi Ren; Zhijie Lin; Zhou Zhao;
707	QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this study, we pioneeringly confirm that properly incorporating activation quantization into the PTQ reconstruction benefits the final accuracy.	Xiuying Wei; Ruihao Gong; Yuhang Li; Xianglong Liu; Fengwei Yu;
708	Quadtree Attention for Vision Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Reduce the computation of transformer by coarse-to-fine attention computation for feature matching/stereo/classification/detection tasks.	Shitao Tang; Jiahui Zhang; Siyu Zhu; Ping Tan;
709	Quantitative Performance Assessment of CNN Units Via Topological Entropy Calculation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel method for quantitatively clarifying the status of individual units in CNNs and show its value in interpreting networks with different generalization ability.	Yang Zhao; Hao Zhang;
710	Query Embedding on Hyper-Relational Knowledge Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We investigate how to extend the multi-hop reasoning problem to hyper-relational queries on knowledge graphs and consider methods for solving it.	Dimitrios Alivanistos; Max Berrendorf; Michael Cochez; Mikhail Galkin;
711	R4D: Utilizing Reference Objects for Long-Range Distance Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We then propose R4D, the first framework to accurately estimate the distance of long-range objects by using references with known distances in the scene.	Yingwei Li; Tiffany Chen; Maya Kabkab; Ruichi Yu; Longlong Jing; Yurong You; Hang Zhao;
712	R5: Rule Discovery with Reinforced and Recurrent Relational Reasoning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we propose R5, a relational reasoning framework based on reinforcement learning that reasons over relational graph data and explicitly mines underlying compositional logical rules from observations.	Shengyao Lu; Bang Liu; Keith G Mills; SHANGLING JUI; Di Niu;
713	Random Matrices in Service of ML Footprint: Ternary Random Features with No Performance Loss Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A novel computational and storage efficient random features technique with no performance loss	Hafiz Tiomoko Ali; Zhenyu Liao; Romain Couillet;
714	Recursive Disentanglement Network Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper has described a solution to the compositional disentangled representation learning problem.	Yixuan Chen; Yubin Shi; Dongsheng Li; Yujiang Wang; Mingzhi Dong; Yingying Zhao; Robert Dick; Qin Lv; Fan Yang; Li Shang;
715	Reducing Excessive Margin to Achieve A Better Accuracy Vs. Robustness Trade-off Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we closely examine the changes induced in the decision boundary of a deep network during adversarial training.	Rahul Rade; Seyed-Mohsen Moosavi-Dezfooli;
716	Reducing The Communication Cost of Federated Learning Through Multistage Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose a multistage optimization scheme that nearly matches the lower bound across all heterogeneity levels.	Charlie Hou; Kiran Koshy Thekumparampil; Giulia Fanti; Sewoong Oh;
717	RegionViT: Regional-to-Local Attention for Vision Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A new architecture for vision transformer	Chun-Fu Chen; Rameswar Panda; Quanfu Fan;
718	Regularized Autoencoders for Isometric Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Regularized Autoencoders that simultaneously learn data manifold and a set of latent space coordinates that preserves the geometry of the learned manifold.	Yonghyeon LEE; Sangwoong Yoon; MinJun Son; Frank C. Park;
719	Reinforcement Learning in Presence of Discrete Markovian Context Evolution Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We consider a context-dependent Reinforcement Learning (RL) setting, which is characterized by: a) an unknown finite number of not directly observable contexts; b) abrupt (discontinuous) context changes; and c) Markovian context evolution.	Hang Ren; Aivar Sootla; Taher Jafferjee; Junxiao Shen; Jun Wang; Haitham Bou Ammar;
720	Reinforcement Learning Under A Multi-agent Predictive State Representation Model: Method and Theory Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new algorithm for MARL under a multi-agent predictive state representation model, where we incorporate a dynamic interaction graph; we provide the theoretical guarantees of our model and run various experiments to support our algorithm.	Zhi Zhang; Zhuoran Yang; Han Liu; Pratap Tokekar; Furong Huang;
721	Reinforcement Learning with Sparse Rewards Using Guidance from Offline Demonstration Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Reinforcement learning in sparse reward environments using offline guidance.	Desik Rengarajan; Gargi Vaidya; Akshay Sarvesh; Dileep Kalathil; Srinivas Shakkottai;
722	Relating Transformers to Models and Neural Representations of The Hippocampal Formation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Transformers learn brain representatations and they are algorithmically related to models of the hippocampal formation.	James C. R. Whittington; Joseph Warren; Tim E.J. Behrens;
723	Relational Multi-Task Learning: Modeling Relations Between Data and Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose MetaLink to solve a variety of multi-task learning settings, by constructing a knowledge graph over data points and tasks.	Kaidi Cao; Jiaxuan You; Jure Leskovec;
724	Relational Surrogate Loss Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we show that directly maintaining the relation of models between surrogate losses and metrics suffices, and propose a rank correlation-based optimization method to maximize this relation and learn surrogate losses.	Tao Huang; Zekang Li; Hua Lu; Yong Shan; Shusheng Yang; Yang Feng; Fei Wang; Shan You; Chang Xu;
725	RelaxLoss: Defending Membership Inference Attacks Without Losing Utility Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel training scheme that is highly effective in protecting against membership inference attacks while preserving the utility of target models.	Dingfan Chen; Ning Yu; Mario Fritz;
726	Reliable Adversarial Distillation with Unreliable Teachers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Therefore, in this paper, we propose reliable introspective adversarial distillation (IAD) where students partially instead of fully trust their teachers.	Jianing Zhu; Jiangchao Yao; Bo Han; Jingfeng Zhang; Tongliang Liu; Gang Niu; Jingren Zhou; Jianliang Xu; Hongxia Yang;
727	RelViT: Concept-guided Vision Transformer for Visual Relational Reasoning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel concept-feature dictionary to enable two new concept-guided auxiliary tasks, which largely improve the model performances on visual relational reasoning, especially for systematic generalization.	Xiaojian Ma; Weili Nie; Zhiding Yu; Huaizu Jiang; Chaowei Xiao; Yuke Zhu; Song-Chun Zhu; Anima Anandkumar;
728	Representation Learning for Online and Offline RL in Low-rank MDPs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study representation learning in low-rank MDP in both online setting and offline setting, and propose statistically and computationally efficient algorithms.	Masatoshi Uehara; Xuezhou Zhang; Wen Sun;
729	Representing Mixtures of Word Embeddings with Mixtures of Topic Embeddings Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A novel method to learn in the word embedding space a set of globally shared topic embedding vectors that are manifested differently in each document	dongsheng wang; Dan dan Guo; He Zhao; Huangjie Zheng; Korawat Tanwisuth; Bo Chen; Mingyuan Zhou;
730	Resonance in Weight Space: Covariate Shift Can Drive Divergence of SGD with Momentum Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that SGDm under covariate shift with fixed step-size can be unstable and diverge due to a phenomenon known as parametric resonance.	Kirby Banman; Garnet Liam Peet-Pare; Nidhi Hegde; Alona Fyshe; Martha White;
731	Responsible Disclosure of Generative Models Using Scalable Fingerprinting Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Our work enables a responsible disclosure of generative models, that allows model inventors to fingerprint their models, so that the generated samples containing a fingerprint can be accurately detected and attributed to a source.	Ning Yu; Vladislav Skripniuk; Dingfan Chen; Larry S. Davis; Mario Fritz;
732	Rethinking Adversarial Transferability from A Data Distribution Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we rethink adversarial transferability from a data distribution perspective and further enhance transferability by score matching based optimization.	Yao Zhu; Jiacheng Sun; Zhenguo Li;
733	Rethinking Class-Prior Estimation for Positive-Unlabeled Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Class-Prior Estimation for Positive-Unlabeled Learning	Yu Yao; Tongliang Liu; Bo Han; Mingming Gong; Gang Niu; Masashi Sugiyama; Dacheng Tao;
734	Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we present a new design for point cloud analysis , dubbed as PointMLP, which delivers new state-of-the-art results on multiple benchmarks and exhibits gratifying inference speed.	Xu Ma; Can Qin; Haoxuan You; Haoxi Ran; Yun Fu;
735	Retriever: Learning Content-Style Representation As A Token-Level Bipartite Graph Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a model-agnostic and unsupervised framework to learn a novel token-level bipartite graph representation of content and style from structured input.	Dacheng Yin; Xuanchi Ren; Chong Luo; Yuwang Wang; Zhiwei Xiong; Wenjun Zeng;
736	Reverse Engineering of Imperceptible Adversarial Image Perturbations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Reverse engineer adversarial image perturbations with a denoiser-based framework.	Yifan Gong; Yuguang Yao; Yize Li; Yimeng Zhang; Xiaoming Liu; Xue Lin; Sijia Liu;
737	Reversible Instance Normalization for Accurate Time-Series Forecasting Against Distribution Shift Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a simple yet effective normalization method, reversible instance normalization (RevIN), which solves the time-series forecasting task against the distribution shift problem.	Taesung Kim; Jinhee Kim; Yunwon Tae; Cheonbok Park; Jang-Ho Choi; Jaegul Choo;
738	Revisiting Design Choices in Offline Model Based Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we compare these heuristics, and design novel protocols to investigate their interaction with other hyperparameters, such as the number of models, or imaginary rollout horizon.	Cong Lu; Philip Ball; Jack Parker-Holder; Michael Osborne; Stephen J. Roberts;
739	Revisiting Flow Generative Models for Out-of-distribution Detection Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we aim to re-examine the potential of generative flow models in OOD detection.	Dihong Jiang; Sun Sun; Yaoliang Yu;
740	Revisiting Over-smoothing in BERT from The Perspective of Graph Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We theoretically analyze the over-smoothing phenomenon of transformer-based models (e.g., BERT) and propose a novel hierarchical fusion strategy to alleviate it.	Han Shi; JIAHUI GAO; Hang Xu; Xiaodan Liang; Zhenguo Li; Lingpeng Kong; Stephen M. S. Lee; James Kwok;
741	Revisit Kernel Pruning with Lottery Regulated Grouped Convolutions Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A simple yet effective structured pruning framework based on kernel pruning, weights shifting, and grouped convolutions.	Shaochen Zhong; Guanqun Zhang; Ningjia Huang; Shuai Xu;
742	Reward Uncertainty for Exploration in Preference-based Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present an exploration method specifically for preference-based RL algorithms.	Xinran Liang; Katherine Shu; Kimin Lee; Pieter Abbeel;
743	Robbing The Fed: Directly Obtaining Private Data in Federated Learning with Modified Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we introduce a new threat model based on minimal but malicious modifications of the shared model architecture which enable the server to directly obtain a verbatim copy of user data from gradient updates without solving difficult inverse problems.	Liam H Fowl; Jonas Geiping; Wojciech Czaja; Micah Goldblum; Tom Goldstein;
744	Robust and Scalable SDE Learning: A Functional Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide an algorithm for estimating the probability of observations of a stochastic process which is significantly faster and more stable than those based on standard integration schemes	Scott Alexander Cameron; Tyron Luke Cameron; Arnu Pretorius; Stephen J. Roberts;
745	Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We leverage proxy distributions to significantly improve the robustness of deep neural network.	Vikash Sehwag; Saeed Mahloujifar; Tinashe Handina; Sihui Dai; Chong Xiang; Mung Chiang; Prateek Mittal;
746	Robust Unlearnable Examples: Protecting Data Privacy Against Adversarial Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes an robust error-minimizing noise that can protect data from being learned under adversarial training.	Shaopeng Fu; Fengxiang He; Yang Liu; Li Shen; Dacheng Tao;
747	RotoGrad: Gradient Homogenization in Multitask Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose an algorithm to simultaneously homogenize gradient magnitudes and directions across tasks in MTL.	Adri�n Javaloy; Isabel Valera;
748	Safe Neurosymbolic Learning with Differentiable Symbolic Execution Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present DSE, the first approach to worst-case-safe parameter learning for potentially non-differentiable neurosymbolic programs where we bridge symbolic execution and stochastic gradient estimator to learn the loss of safety properties.	Chenxi Yang; Swarat Chaudhuri;
749	Sample and Computation Redistribution for Efficient Face Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We search for optimised computation distribution and training sample distribution for the task of face detection.	Jia Guo; Jiankang Deng; Alexandros Lattas; Stefanos Zafeiriou;
750	Sample Efficient Deep Reinforcement Learning Via Uncertainty Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: The sample efficiency and performance of model-free DRL is improved by estimating the predictive uncertainty of the targets using probabilistic ensembles and down-weighting the uncertain samples using batch inverse-variance weighting.	Vincent Mai; Kaustubh Mani; Liam Paull;
751	Sample Efficient Stochastic Policy Extragradient Algorithm for Zero-Sum Markov Game Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes a fully decentralized, model-free, provably convergent, sample efficient stochastic policy extragradient algorithm with symmetric and private policy updates	Ziyi Chen; Shaocong Ma; Yi Zhou;
752	Sample Selection with Uncertainty of Losses for Learning with Noisy Labels Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we incorporate the uncertainty of losses by adopting interval estimation instead of point estimation of losses, where lower bounds of the con?dence intervals of losses derived from distribution-free concentration inequalities, but not losses themselves, are used for sample selection.	Xiaobo Xia; Tongliang Liu; Bo Han; Mingming Gong; Jun Yu; Gang Niu; Masashi Sugiyama;
753	Sampling with Mirrored Stein Operators Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce multi-particle generalization of mirror descent for sampling in constrained domains and non-Euclidean geometries.	Jiaxin Shi; Chang Liu; Lester Mackey;
754	Scalable One-Pass Optimisation of High-Dimensional Weight-Update Hyperparameters By Implicit Differentiation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We develop a gradient-based hyperparameter optimisation algorithm, applicable to a wide range of continuous hyperparameters, and scaling to large numbers of hyperparameters, without dramatically increasing training time from the non-HPO baseline.	Ross M Clarke; Elre Talea Oldewage; Jos� Miguel Hern�ndez-Lobato;
755	Scalable Sampling for Nonsymmetric Determinantal Point Processes Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the first scalable linear-time and sublinear-time sampling algorithms for nonsymmetric determinantal point processes.	Insu Han; Mike Gartrell; Jennifer Gillenwater; Elvis Dohmatob; amin karbasi;
756	Scale Efficiently: Insights from Pretraining and Finetuning Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Scaling laws for upstream and downstream tasks	Yi Tay; Mostafa Dehghani; Jinfeng Rao; William Fedus; Samira Abnar; Hyung Won Chung; Sharan Narang; Dani Yogatama; Ashish Vaswani; Donald Metzler;
757	Scale Mixtures of Neural Network Gaussian Processes Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Infinitely-wide neural networks can be equivalent to scale mixtures of Gaussian processes.	Hyungi Lee; Eunggu Yun; Hongseok Yang; Juho Lee;
758	Scaling Laws for Neural Machine Translation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide (model) scaling laws for neural machine translation.	Behrooz Ghorbani; Orhan Firat; Markus Freitag; Ankur Bapna; Maxim Krikun; Xavier Garcia; Ciprian Chelba; Colin Cherry;
759	Scaling The Depth of Vision Transformers Via The Fourier Domain Analysis Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we investigate the scalability issue with ViT via Fourier domain analysis and propose two practical solutions by scaling different frequency components of attention and feature maps.	Peihao Wang; Wenqing Zheng; Tianlong Chen; Zhangyang Wang;
760	Scarf: Self-Supervised Contrastive Learning Using Random Feature Corruption Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Scarf is a self-supervised, contrastive pre-training method for neural networks applied to tabular classification tasks that boosts performance, even when labeled data is limited or noisy.	Dara Bahri; Heinrich Jiang; Yi Tay; Donald Metzler;
761	Scattering Networks on The Sphere for Scalable and Rotationally Equivariant Spherical CNNs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Scaling rotationally equivariant spherical CNNs to high-resolution data through spherical scattering networks	Jason McEwen; Christopher Wallis; Augustine N. Mavor-Parker;
762	Scene Transformer: A Unified Architecture for Predicting Future Trajectories of Multiple Agents Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce a scene-centric masked sequence based motion prediction model that unifies a variety of motion prediction tasks from joint motion predictions to conditioned prediction.	Jiquan Ngiam; Vijay Vasudevan; Benjamin Caine; Zhengdong Zhang; Hao-Tien Lewis Chiang; Jeffrey Ling; Rebecca Roelofs; Alex Bewley; Chenxi Liu; Ashish Venugopal; David J Weiss; Ben Sapp; Zhifeng Chen; Jonathon Shlens;
763	Score-Based Generative Modeling with Critically-Damped Langevin Diffusion Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we propose a novel diffusion process ideally suited for score-based generative models and provide new insights into score-based denoising diffusion models.	Tim Dockhorn; Arash Vahdat; Karsten Kreis;
764	SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To address these issues, we introduce a new image synthesis and editing method, Stochastic Differential Editing (SDEdit), based on a diffusion model generative prior, which synthesizes realistic images by iteratively denoising through a stochastic differential equation (SDE).	Chenlin Meng; Yutong He; Yang Song; Jiaming Song; Jiajun Wu; Jun-Yan Zhu; Stefano Ermon;
765	Selective Ensembles for Consistent Predictions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Deep models give inconsistent predictions and explanations over small changes (e.g. random initialization). We can mitigate this by using selective ensemble models, which abstain from prediction if their constituent models do not agree sufficiently.	Emily Black; Klas Leino; Matt Fredrikson;
766	Self-ensemble Adversarial Training for Improved Robustness Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper proposes an efficient self-ensemble method for adversarial trained classifiers and significantly improve their adversarial robustness	Hongjun Wang; Yisen Wang;
767	Self-Joint Supervised Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To address these limitations, we propose a new supervised learning paradigm called self-joint learning that generalizes the standard approach by modeling the joint conditional distribution of two observed samples, where each sample is an image and its label.	Navid Kardan; Mubarak Shah; Mitch Hill;
768	Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Self-supervised graph neural networks for seizure detection and classification from EEG.	Siyi Tang; Jared Dunnmon; Khaled Kamal Saab; Xuan Zhang; Qianying Huang; Florian Dubost; Daniel Rubin; Christopher Lee-Messer;
769	Self-supervised Learning Is More Robust to Dataset Imbalance Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show that self-supervised pre-training yields representations more robust to dataset imbalance, because it captures more diverse features from the frequent classes, and can be improved further by re-weighting regularization.	Hong Liu; Jeff Z. HaoChen; Adrien Gaidon; Tengyu Ma;
770	Self-Supervision Enhanced Feature Selection with Correlated Gates Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel DL-based feature selection method using self-supervised learning and multivariate Bernoulli distribution to address common challenges in feature selection: a scarcity of labeled samples and significant correlations among features.	Changhee Lee; Fergus Imrie; Mihaela van der Schaar;
771	Semi-relaxed Gromov-Wasserstein Divergence and Applications on Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A new transport based divergence between structured data induced by the relaxation of a mass constraint of the Gromov-Wasserstein problem, leading to new SOTA performances for unsupervised ML applications on graphs.	C�dric Vincent-Cuaz; R�mi Flamary; Marco Corneli; Titouan Vayer; Nicolas Courty;
772	Sequence Approximation Using Feedforward Spiking Neural Network for Spatiotemporal Learning: Theory and Optimization Methods Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A theoretical approache to study the approximation capability of feedforward spiking neural network and optimization methods for such network.	Xueyuan She; Saurabh Dash; Saibal Mukhopadhyay;
773	Sequential Reptile: Inter-Task Gradient Alignment for Multilingual Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a simple yet effective gradient alignment method for finetuning multilingual pretrained language models.	Seanie Lee; Hae Beom Lee; Juho Lee; Sung Ju Hwang;
774	SGD Can Converge to Local Maxima Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that it can be common for SGD to converge to saddle points and maxima.	Liu Ziyin; Botao Li; James B Simon; Masahito Ueda;
775	Shallow and Deep Networks Are Near-Optimal Approximators of Korobov Functions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We analyze the number of neurons and training parameters that a neural network needs to approximate multivariate functions of bounded second mixed derivatives	Moise Blanchard; Mohammed Amine Bennouna;
776	SHINE: SHaring The INverse Estimate from The Forward Pass for Bi-level Optimization and Implicit Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Use the approximate Jacobian matrix computed in quasi-Newton methods to perform the inversion needed in the training of implicit models.	Zaccharie Ramzi; Florian Mannel; Shaojie Bai; Jean-Luc Starck; Philippe Ciuciu; Thomas Moreau;
777	Should I Run Offline Reinforcement Learning or Behavioral Cloning? Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Characterization of scenarios where offline reinforcement learning outperforms behavioral cloning	Aviral Kumar; Joey Hong; Anikait Singh; Sergey Levine;
778	Should We Be Pre-training? An Argument for End-task Aware Training As An Alternative Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: When we know the end-task objective in advance, instead of pre-training on auxiliary objectives and then fine-tuning, we advocate for it to be introduced directly in training with auxiliary objectives	Lucio M. Dery; Paul Michel; Ameet Talwalkar; Graham Neubig;
779	Shuffle Private Stochastic Convex Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The first analysis of shuffle private stochastic convex optimization.	Albert Cheu; Matthew Joseph; Jieming Mao; Binghui Peng;
780	Simple GNN Regularisation for 3D Molecular Property Prediction and Beyond Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A simple regularisation technique for GNNs applied to 3D molecular property prediction & beyond.	Jonathan Godwin; Michael Schaarschmidt; Alexander L Gaunt; Alvaro Sanchez-Gonzalez; Yulia Rubanova; Petar Velickovic; James Kirkpatrick; Peter Battaglia;
781	SimVLM: Simple Visual Language Model Pretraining with Weak Supervision Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this work, we relax these constraints and present a minimalist pretraining framework, named Simple Visual Language Model (SimVLM).	Zirui Wang; Jiahui Yu; Adams Wei Yu; Zihang Dai; Yulia Tsvetkov; Yuan Cao;
782	SketchODE: Learning Neural Sketch Representation in Continuous Time Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Modelling continuous time chirographic structures like handwriting, diagrams, sketches etc with Neural Ordinary Differential Equations.	Ayan Das; Yongxin Yang; Timothy Hospedales; Tao Xiang; Yi-Zhe Song;
783	Skill-based Meta-Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we devise a method that enables meta-learning on long-horizon, sparse-reward tasks, allowing us to solve unseen target tasks with orders of magnitude fewer environment interactions.	Taewook Nam; Shao-Hua Sun; Karl Pertsch; Sung Ju Hwang; Joseph J Lim;
784	Solving Inverse Problems in Medical Imaging with Score-Based Generative Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To address this issue, we propose a fully unsupervised technique for inverse problem solving, leveraging the recently introduced score-based generative models.	Yang Song; Liyue Shen; Lei Xing; Stefano Ermon;
785	SOSP: Efficiently Capturing Global Correlations By Second-Order Structured Pruning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a second-order structured pruning method which efficiently captures global correlations among structures of deep neural networks.	Manuel Nonnenmacher; Thomas Pfeil; Ingo Steinwart; David Reeb;
786	Sound Adversarial Audio-Visual Navigation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work aims to do an adversarial sound intervention for robust audio-visual navigation.	Yinfeng Yu; Wenbing Huang; Fuchun Sun; Changan Chen; Yikai Wang; Xiaohong Liu;
787	Sound and Complete Neural Network Repair with Minimality and Locality Guarantees Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a novel methodology for repairing neural networks that use ReLU activation functions.	Feisi Fu; Wenchao Li;
788	Source-Free Adaptation to Measurement Shift Via Bottom-Up Feature Restoration Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We identify a type of domain shift which can be resolved by restoring the same features and address it in the source-free setting by using softly-binned histograms to cheaply and flexibly align the marginal feature distributions.	Cian Eastwood; Ian Mason; Chris Williams; Bernhard Sch�lkopf;
789	Space-Time Graph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce space-time graph neural network (ST-GNN) tailored to jointly process the underlying space-time topology of time-varying network data, and we show its stability to perturbations.	Samar Hadou; Charilaos I Kanatsoulis; Alejandro Ribeiro;
790	Spanning Tree-based Graph Generation for Molecules Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new molecular graph generative model based on compact tree constructive operators.	Sungsoo Ahn; Binghong Chen; Tianzhe Wang; Le Song;
791	Sparse Attention with Learning to Hash Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a new strategy for sparse attention, namely LHA (Learning-to-Hash Attention), which directly learns separate parameterized hash functions for queries and keys, respectively.	Zhiqing Sun; Yiming Yang; Shinjae Yoo;
792	Sparse DETR: Efficient End-to-End Object Detection with Learnable Sparsity Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Sparse DETR is an efficient end-to-end object detector that sparsifies encoder queries by using the learnable decoder attention map predictor. It achieves better performance than Deformable DETR even with only 10% encoder queries on the COCO dataset.	Byungseok Roh; JaeWoong Shin; Wuhyun Shin; Saehoon Kim;
793	Sparsity Winning Twice: Better Robust Generalization from More Efficient Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we investigate this intriguing problem from a new perspective, i.e., $\textit{injecting appropriate forms of sparsity}$ during adversarial training.	Tianlong Chen; Zhenyu Zhang; pengjun wang; Santosh Balachandra; Haoyu Ma; Zehao Wang; Zhangyang Wang;
794	Spatial Graph Attention and Curiosity-driven Policy for Antiviral Drug Discovery Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We developed a reinforcement learning framework that advances in exploiting spatial and attributional molecular information as well as exploring novel and synthesizable chemical structures for the purpose of antiviral drug discovery.	Yulun Wu; Nicholas Choma; Andrew Deru Chen; Mikaela Cashman; Erica Teixeira Prates; Veronica G Melesse Vergara; Manesh B Shah; Austin Clyde; Thomas Brettin; Wibe Albert de Jong; Neeraj Kumar; Martha S Head; Rick L. Stevens; Peter Nugent; Daniel A Jacobson; James B Brown;
795	SphereFace2: Binary Classification Is All You Need for Deep Face Recognition Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A novel deep face recognition framework	Yandong Wen; Weiyang Liu; Adrian Weller; Bhiksha Raj; Rita Singh;
796	Spherical Message Passing for 3D Molecular Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we conduct analyses in the spherical coordinate system (SCS) for the complete identification of 3D graph structures.	Yi Liu; Limei Wang; Meng Liu; Yuchao Lin; Xuan Zhang; Bora Oztekin; Shuiwang Ji;
797	Spike-inspired Rank Coding for Fast and Accurate Recurrent Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Learning to infer fast in LSTMs inspired by SNNs, and applied in speech recognition	Alan Jeffares; Qinghai Guo; Pontus Stenetorp; Timoleon Moraitis;
798	Spread Spurious Attribute: Improving Worst-group Accuracy with Spurious Attribute Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Using a small amount of attribute annotated samples for training can boost worst-group performance in the presence of spurious correlation.	Junhyun Nam; Jaehyung Kim; Jaeho Lee; Jinwoo Shin;
799	Sqrt(d) Dimension Dependence of Langevin Monte Carlo Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: The known dimension dependence of LMC is improved, under regularity assumptions, from d to sqrt(d), based on a refined mean square analysis framework.	Ruilin Li; Hongyuan Zha; Molei Tao;
800	SQuant: On-the-Fly Data-Free Quantization Via Diagonal Hessian Approximation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A fast and accurate data-free quantization framework named SQuant.	Cong Guo; Yuxian Qiu; Jingwen Leng; Xiaotian Gao; Chen Zhang; Yunxin Liu; Fan Yang; Yuhao Zhu; Minyi Guo;
801	Stability Regularization for Discrete Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a method for training neural network models with discrete stochastic variables.	Adeel Pervez; Efstratios Gavves;
802	Steerable Partial Differential Operators for Equivariant Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present a framework for equivariant partial differential operators, generalizing existing approaches and narrowing the gap between PDOs and convolutions.	Erik Jenner; Maurice Weiler;
803	Stein Latent Optimization for Generative Adversarial Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel GAN method that performs unsupervised conditional generation robustly on real-world datasets with balanced or imbalanced attributes even in the absence of attribute information (e.g., the imbalance ratio)	Uiwon Hwang; Heeseung Kim; Dahuin Jung; Hyemi Jang; Hyungyu Lee; Sungroh Yoon;
804	Step-unrolled Denoising Autoencoders for Text Generation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new generative model of text that unrolls the denoising process during training.	Nikolay Savinov; Junyoung Chung; Mikolaj Binkowski; Erich Elsen; Aaron van den Oord;
805	Stiffness-aware Neural Network for Learning Hamiltonian Systems Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose stiffness-aware neural network (SANN), a new method for learning Hamiltonian dynamical systems from data.	SENWEI Liang; Zhongzhan Huang; Hong Zhang;
806	Stochastic Training Is Not Necessary for Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Models trained with full-batch gradient descent and explicit regularization can match the generalization performance of models trained with stochastic minibatching.	Jonas Geiping; Micah Goldblum; Phil Pope; Michael Moeller; Tom Goldstein;
807	Strength of Minibatch Noise in SGD Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We solve the strength and shape of the minibatch noise in SGD exactly.	Liu Ziyin; Kangqiao Liu; Takashi Mori; Masahito Ueda;
808	Structure-Aware Transformer Policy for Inhomogeneous Multi-Task Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present a modular Multi-task Reinforcement Learning method for inhomogeneous control tasks incorporating structural embedding of morphology.	Sunghoon Hong; Deunsol Yoon; Kee-Eung Kim;
809	StyleNeRF: A Style-based 3D Aware Generator for High-resolution Image Synthesis Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present StyleNeRF, a 3D-aware generative model that can synthesize high-resolution images with high multi-view consistency.	Jiatao Gu; Lingjie Liu; Peng Wang; Christian Theobalt;
810	Subspace Regularizers for Few-Shot Class Incremental Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a simple yet highly effective set of subspace-based regularizers to address representation learning for few-shot incremental classification.	Afra Feyza Aky�rek; Ekin Aky�rek; Derry Wijaya; Jacob Andreas;
811	SUMNAS: Supernet with Unbiased Meta-Features for Neural Architecture Search Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a supernet learning strategy that learns unbiased meta-features to tackle multi-model forgetting problem of neural architecture search.	Hyeonmin Ha; Ji-Hoon Kim; Semin Park; Byung-Gon Chun;
812	Superclass-Conditional Gaussian Mixture Model For Learning Fine-Grained Embeddings Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a training framework characterized by a novel superclass conditional Gaussian mixture (SCGM) based generative model for learning fine-grained representations for cross-granularity adaptation.	Jingchao Ni; Wei Cheng; Zhengzhang Chen; Takayoshi Asakura; Tomoya Soma; Sho Kato; Haifeng Chen;
813	Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work proposes a novel training paradigm, Data efficient CLIP (DeCLIP), to alleviate this limitation.	Yangguang Li; Feng Liang; Lichen Zhao; Yufeng Cui; Wanli Ouyang; Jing Shao; Fengwei Yu; Junjie Yan;
814	SURF: Semi-supervised Reward Learning with Data Augmentation for Feedback-efficient Preference-based Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present SURF, a semi-supervised reward learning algorithm with data augmentation for feedback-efficient preference-based RL.	Jongjin Park; Younggyo Seo; Jinwoo Shin; Honglak Lee; Pieter Abbeel; Kimin Lee;
815	Surreal-GAN:Semi-Supervised Representation Learning Via GAN for Uncovering Heterogeneous Disease-related Imaging Patterns Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We proposed a novel method, Surreal-GAN, to derive low dimensional representation of disease-related patterns from neuroimaging data.	Zhijian Yang; Junhao Wen; Christos Davatzikos;
816	Surrogate NAS Benchmarks: Going Beyond The Limited Search Spaces of Tabular NAS Benchmarks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We present surrogate benchmarks for neural architecture search and a general methodology for constructing them.	Arber Zela; Julien Niklas Siems; Lucas Zimmer; Jovita Lukasik; Margret Keuper; Frank Hutter;
817	Switch to Generalize: Domain-Switch Learning for Cross-Domain Few-Shot Classification Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper considers few-shot learning under the cross-domain scenario.	Zhengdong Hu; Yifan Sun; Yi Yang;
818	Symbolic Learning to Optimize: Towards Interpretability and Scalability Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Learning to distill learned optimization rule into symbolic math equations that bears better interpretability and scales better.	Wenqing Zheng; Tianlong Chen; Ting-Kuei Hu; Zhangyang Wang;
819	Synchromesh: Reliable Code Generation from Pre-trained Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A framework to generate programs from large pre-trained language models (e.g. GPT-3, Codex) while satisfying syntactic and semantic constraints.	Gabriel Poesia; Alex Polozov; Vu Le; Ashish Tiwari; Gustavo Soares; Christopher Meek; Sumit Gulwani;
820	Tackling The Generative Learning Trilemma with Denoising Diffusion GANs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: To reduce the number of sampling steps in diffusion models, we propose to model the denoising distribution with conditional GANs. We show our model tackles the generative learning trilemma & achieves high sample quality, diversity & fast sampling.	Zhisheng Xiao; Karsten Kreis; Arash Vahdat;
821	TAda! Temporally-Adaptive Convolutions for Video Understanding Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A stand-alone temporal modelling module or a plug-in enhancement of the 1D/2D/3D convolutions used in video models for better and more efficient temporal modelling.	Ziyuan Huang; Shiwei Zhang; Liang Pan; Zhiwu Qing; Mingqian Tang; Ziwei Liu; Marcelo H Ang Jr;
822	Taming Sparsely Activated Transformer with Stochastic Experts Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a new variant of MoE, Transformer with Stochastic Experts, that is more parameter efficient.	Simiao Zuo; Xiaodong Liu; Jian Jiao; Young Jin Kim; Hany Hassan; Ruofei Zhang; Jianfeng Gao; Tuo Zhao;
823	TAMP-S2GCNets: Coupling Time-Aware Multipersistence Knowledge Representation with Spatio-Supra Graph Convolutional Networks for Time-Series Forecasting Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We make the first step toward integrating two emerging directions, time-aware deep learning and multi-parameter persistence, allowing us to infer latent time-conditioned relations among entities in multivariate time series forecasting tasks.	Yuzhou Chen; Ignacio Segovia-Dominguez; Baris Coskunuzer; Yulia Gel;
824	TAPEX: Table Pre-training Via Learning A Neural SQL Executor Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work performs table pre-training by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries and their execution results.	Qian Liu; Bei Chen; Jiaqi Guo; Morteza Ziyadi; Zeqi Lin; Weizhu Chen; Jian-Guang Lou;
825	Target-Side Data Augmentation for Sequence Generation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study the data augmentation for target-side conditional input of autoregressive sequence generation and propose a new method to build soft synthetic data.	Shufang Xie; Ang Lv; Yingce Xia; Lijun Wu; Tao Qin; Rui Yan; Tie-Yan Liu;
826	Task Affinity with Maximum Bipartite Matching in Few-Shot Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Task affinity and its application in few-shot learning	Cat Phuoc Le; Juncheng Dong; Mohammadreza Soltani; Vahid Tarokh;
827	Task-Induced Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce task-induced representation learning, which leverages task information in offline data from prior tasks to learn representations of visually complex scenes that model only task-relevant aspects and enable efficient learning of new tasks.	Jun Yamada; Karl Pertsch; Anisha Gunjal; Joseph J Lim;
828	Task Relatedness-Based Generalization Bounds for Meta Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this paper, we propose to address this problem by defining a new notion of task relatedness according to the existence of the bijective transformation between two tasks.	Jiechao Guan; Zhiwu Lu;
829	Temporal Alignment Prediction for Supervised Representation Learning and Few-Shot Sequence Classification Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a learnable sequence distance by predicting the temporal alignment and show its application in supervised representation learning for sequence data and few-shot action recognition.	Bing Su; Ji-Rong Wen;
830	Temporal Efficient Training of Spiking Neural Network Via Gradient Re-weighting Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper provides a novel temporal efficient training method for SNN, which significantly improves performance by modifying the optimization target.	Shikuang Deng; Yuhang Li; Shanghang Zhang; Shi Gu;
831	The Boltzmann Policy Distribution: Accounting for Systematic Suboptimality in Human Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose modeling human behavior with a Boltzmann distribution over policies?not trajectories?and show it is more accurate and useful.	Cassidy Laidlaw; Anca Dragan;
832	The Convex Geometry of Backpropagation: Neural Network Gradient Flows Converge to Extreme Points of The Dual Convex Program Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study non-convex subgradient flows for training two-layer ReLU neural networks from a convex geometry and duality perspective.	Yifei Wang; Mert Pilanci;
833	The Effects of Invertibility on The Representational Complexity of Encoders in Variational Autoencoders Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: VAEs with invertible mean map have small approximate encoders; non-invertible maps can result in large encoders.	Divyansh Pareek; Andrej Risteski;
834	The Essential Elements of Offline RL Via Supervised Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Experimentally evaluating when and why supervised learning solves offline RL	Scott Emmons; Benjamin Eysenbach; Ilya Kostrikov; Sergey Levine;
835	The Evolution of Uncertainty of Learning in Games Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that the differential entropy of certain learning-in-game systems increases linearly with time, formalizing their increased unpredictability over time.	Yun Kuen Cheung; Georgios Piliouras; Yixin Tao;
836	The Geometry of Memoryless Stochastic Policy Optimization in Infinite-Horizon POMDPs Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We provide an explicit description of the optimization problem and derive bounds on the number of critical points in POMDPs with memoryless stochastic policies depending on the degree of observability.	Johannes M�ller; Guido Montufar;
837	The Inductive Bias of In-Context Learning: Rethinking Pretraining Example Design Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We prove that pertained LMs model stronger dependencies between sentences that were shown in same training example, thus indicating benefits of better informed pretraining example design	Yoav Levine; Noam Wies; Daniel Jannai; Dan Navon; Yedid Hoshen; Amnon Shashua;
838	The MultiBERTs: BERT Reproductions for Robustness Analysis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce MultiBERTs, 25 BERT checkpoints trained with similar hyper-parameters but different random seeds, and the Multi-Bootstrap, a bootstrapping method for experimental settings that involve multiple models and limited test data.	Thibault Sellam; Steve Yadlowsky; Ian Tenney; Jason Wei; Naomi Saphra; Alexander D’Amour; Tal Linzen; Jasmijn Bastings; Iulia Raluca Turc; Jacob Eisenstein; Dipanjan Das; Ellie Pavlick;
839	The Rich Get Richer: Disparate Impact of Semi-Supervised Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We reveal the disparate impacts of deploying SSL: the rich sub-population (higher baseline accuracy without SSL) benefits more from SSL; while the poor sub-population (low baseline accuracy) might even observe a performance drop after SSL.	Zhaowei Zhu; Tianyi Luo; Yang Liu;
840	The Role of Pretrained Representations for The OOD Generalization of RL Agents Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study the role of pretrained representations for the out-of-distribution generalization of RL agents.	Frederik Tr�uble; Andrea Dittadi; Manuel Wuthrich; Felix Widmaier; Peter Vincent Gehler; Ole Winther; Francesco Locatello; Olivier Bachem; Bernhard Sch�lkopf; Stefan Bauer;
841	The Spectral Bias of Polynomial Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study the spectral bias of polynomial networks and compare it with the spectral bias of standard neural nets using kernel approximations	Moulik Choraria; Leello Tadesse Dadi; Grigorios Chrysos; Julien Mairal; Volkan Cevher;
842	The Three Stages of Learning Dynamics in High-dimensional Kernel Methods Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We study the training dynamics of gradient flows on the population risk and empirical risk of high-dimensional kernel least-squares problems, which we show has three learning stages.	Nikhil Ghosh; Song Mei; Bin Yu;
843	The Uncanny Similarity of Recurrence and Depth Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show quantitatively and qualitatively that recurrent models have the same behaviors as feed-forward networks despite reusing parameters at each recurrence.	Avi Schwarzschild; Arjun Gupta; Amin Ghiasi; Micah Goldblum; Tom Goldstein;
844	The Unreasonable Effectiveness of Random Pruning: Return of The Most Naive Baseline for Sparse Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We revisit the most naive baseline, random pruning, in sparse training and highlight a perhaps counter-intuitive finding: random pruning can be quite powerful for the sparse training of modern neural networks.	Shiwei Liu; Tianlong Chen; Xiaohan Chen; Li Shen; Decebal Constantin Mocanu; Zhangyang Wang; Mykola Pechenizkiy;
845	THOMAS: Trajectory Heatmap Output with Learned Multi-Agent Sampling Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a solution for multi-agent coherent multimodal trajectory prediction by learning a recombination of each agent predicted modalities.	Thomas Gilles; Stefano Sabatini; Dzmitry Tsishkou; Bogdan Stanciulescu; Fabien Moutarde;
846	Tighter Sparse Approximation Bounds for ReLU Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show conditions under which a function can be represented by an infinite-width neural network on a bounded set, and refine sparse neural network approximation bounds.	Carles Domingo-Enrich; Youssef Mroueh;
847	Top-label Calibration and Multiclass-to-binary Reductions Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose top-label calibration, a new and arguably natural notion for multiclass calibration, along with ‘wrapper’ calibration algorithms that reduce multiclass calibration to binary calibration.	Chirag Gupta; Aaditya Ramdas;
848	Top-N: Equivariant Set and Graph Generation Without Exchangeability Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the Top-N method for one-shot set and graph probabilistic decoders as a replacement for i.i.d. generation in the first layer.	Clement Vignac; Pascal Frossard;
849	Topological Graph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We describe a new layer for graph neural networks that incorporates multi-scale (ranging from local to global) topological information.	Max Horn; Edward De Brouwer; Michael Moor; Yves Moreau; Bastian Rieck; Karsten Borgwardt;
850	Topologically Regularized Data Embeddings Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A method for incorporating expert prior topological knowledge into data embeddings.	Robin Vandaele; Bo Kang; Jefrey Lijffijt; Tijl De Bie; Yvan Saeys;
851	Toward Efficient Low-Precision Training: Data Format Optimization and Hysteresis Quantization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a systematic data format optimization method and hysteresis quantization scheme to enable efficient low-precision training.	Sunwoo Lee; Jeongwoo Park; Dongsuk Jeon;
852	Toward Faithful Case-based Reasoning Through Learning Prototypes in A Nearest Neighbor-friendly Space Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Offering better prototype explanations using a nearest-neighbor friendly embedding space.	Seyed Omid Davoudi; Majid Komeili;
853	Towards A Unified View of Parameter-Efficient Transfer Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a unified framework for several state-of-the-art parameter-efficient tuning methods,	Junxian He; Chunting Zhou; Xuezhe Ma; Taylor Berg-Kirkpatrick; Graham Neubig;
854	Towards Better Understanding and Better Generalization of Low-shot Classification in Histology Images with Contrastive Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Here, we pioneer the study of low-shot learning for histology images by setting up three cross-domain tasks that reflect real clinics problems.	Jiawei Yang; Hanbo Chen; Jiangpeng Yan; Xiaoyu Chen; Jianhua Yao;
855	Towards Continual Knowledge Learning of Language Models Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel continual learning formulation named Continual Knowledge Learning which allows large language models to constantly obtain new and updated knowledge while mitigating forgetting of previous learned time-invariant knowledge. We construct a new benchmark and metric to quantify the retention of time-invariant world knowledge, the update of outdated knowledge, and the acquisition of new knowledge.	Joel Jang; Seonghyeon Ye; Sohee Yang; Joongbo Shin; Janghoon Han; Gyeonghun KIM; Stanley Jungkyu Choi; Minjoon Seo;
856	Towards Deployment-Efficient Reinforcement Learning: Lower Bound and Optimality Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a formal theoretical formulation for depolyment-efficient reinforcement learning; establish lower bounds for deployment complexity and study near-optimal deployment-efficient algorithms in linear MDP setting.	Jiawei Huang; Jinglin Chen; Li Zhao; Tao Qin; Nan Jiang; Tie-Yan Liu;
857	Towards Distribution Shift of Node-Level Prediction on Graphs: An Invariance Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We formulate out-of-distribution generalization problem for node-level prediction on graphs and propose a new learning approach based on invariant models	Qitian Wu; Hengrui Zhang; Junchi Yan; David Wipf;
858	Towards Evaluating The Robustness of Neural Networks Learned By Transduction Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Exploring evaluating the adversarial robustness of transductive-learning based defenses.	Jiefeng Chen; Xi Wu; Yang Guo; Yingyu Liang; Somesh Jha;
859	Towards General Function Approximation in Zero-Sum Markov Games Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In the decoupled setting where the agent controls a single player and plays against an arbitrary opponent, we propose a new model-free algorithm.	Baihe Huang; Jason D. Lee; Zhaoran Wang; Zhuoran Yang;
860	Towards Model Agnostic Federated Learning Using Knowledge Distillation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We develop a rich yet tractable framework for analyzing distillation based federated learning algorithms, using which we draw some surprising insights.	Andrei Afonin; Sai Praneeth Karimireddy;
861	Towards Training Billion Parameter Graph Neural Networks for Atomic Simulations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We scale GNNs used for modeling atomic simulations by an order of magnitude and obtain large performance improvements on the Open Catalyst 2020 dataset.	Anuroop Sriram; Abhishek Das; Brandon M Wood; C. Lawrence Zitnick;
862	Towards Understanding The Data Dependency of Mixup-style Training Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A theoretical analysis of data conditions under which mixup can perform worse, better, and identically when compared to empirical risk minimization.	Muthu Chidambaram; Xiang Wang; Yuzheng Hu; Chenwei Wu; Rong Ge;
863	Towards Understanding The Robustness Against Evasion Attack on Categorical Data Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper explores the characterization and certification of the robustness against evasion attack with categorical input.	Hongyan Bao; Yufei Han; Yujun Zhou; Yun Shen; Xiangliang Zhang;
864	TPU-GAN: Learning Temporal Coherence from Dynamic Point Cloud Sequences Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a GAN framework for super-resolution task on dynamic point cloud sequences.	Zijie Li; Tianqin Li; Amir Barati Farimani;
865	Tracking The Risk of A Deployed Model and Detecting Harmful Distribution Shifts Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we design simple sequential tools for testing if the difference between source (training) and target (test) distributions leads to a significant drop in a risk function of interest, like accuracy or calibration.	Aleksandr Podkopaev; Aaditya Ramdas;
866	TRAIL: Near-Optimal Imitation Learning with Suboptimal Data Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: A provably beneficial way to learn action representations for imitation learning from suboptimal auxiliary data.	Mengjiao Yang; Sergey Levine; Ofir Nachum;
867	Training Data Generating Networks: Shape Reconstruction Via Bi-level Optimization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel 3d shape representation for 3d shape reconstruction from a single image.	Biao Zhang; Peter Wonka;
868	Training Invariances and The Low-rank Phenomenon: Beyond Linear Networks Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We extend theoretical results regarding the low-rank bias of deep linear neural networks trained with gradient-based algorithm to non-linear architectures, reflecting empirical results in the literature.	Thien Le; Stefanie Jegelka;
869	Training Transition Policies Via Distribution Matching for Complex Tasks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Training transition policies via distribution matching	JU-SEUNG BYUN; Andrew Perrault;
870	Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We show that our simple position method enables transformer LMs to efficiently and accurately perform inference on longer sequences than they were trained on.	Ofir Press; Noah Smith; Mike Lewis;
871	Trans-Encoder: Unsupervised Sentence-pair Modelling Through Self- and Mutual-distillations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Bootstrapping an unsupervised sentence encoder by self-distilling knowledge between its bi-encoder and cross-encoder forms, enhancing each other iteratively.	Fangyu Liu; Yunlong Jiao; Jordan Massiah; Emine Yilmaz; Serhii Havrylov;
872	Transferable Visual Control Policies Through Robot-Awareness Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We closely integrate readily available knowledge about the robot and world into a learned model to facilitate transfer.	Edward S. Hu; Kun Huang; Oleh Rybkin; Dinesh Jayaraman;
873	Transfer RL Across Observation Feature Spaces Via Model-Based Regularization Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a model-based transfer learning algorithm that transfers knowledge across tasks with drastically different observation spaces, without any prior knowledge of the inter-task mapping.	Yanchao Sun; Ruijie Zheng; Xiyao Wang; Andrew E Cohen; Furong Huang;
874	Transformer-based Transform Coding Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Specifically, we show that nonlinear transforms built on Swin-transformers can achieve better compression efficiency than transforms built on convolutional neural networks (ConvNets), while requiring fewer parameters and shorter decoding time.	Yinhao Zhu; Yang Yang; Taco Cohen;
875	Transition to Linearity of Wide Neural Networks Is An Emerging Property of Assembling Weak Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Transition to linearity of wide neural networks is an emerging property of assembling weak models corresponding to individual neurons	Chaoyue Liu; Libin Zhu; Misha Belkin;
876	TRGP: Trust Region Gradient Projection for Continual Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a novel continual learning approach to facilitate the forward knowledge transfer, based on an efficient characterization of task correlation using a novel notion of ‘trust region’.	Sen Lin; Li Yang; Deliang Fan; Junshan Zhang;
877	Triangle and Four Cycle Counting with Predictions in Graph Streams Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose algorithms for counting triangles and four cycles in graph streams with the aid of ML predictors.	Justin Y Chen; Talya Eden; Piotr Indyk; Honghao Lin; Shyam Narayanan; Ronitt Rubinfeld; Sandeep Silwal; Tal Wagner; David Woodruff; Michael Zhang;
878	Trigger Hunting with A Topological Prior for Trojan Detection Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: To this end, we propose innovative priors such as \textbf{diversity} and \textbf{topological simplicity} to not only increase the chances of finding the appropriate triggers but also improve the quality of the found triggers.	Xiaoling Hu; Xiao Lin; Michael Cogswell; Yi Yao; Susmit Jha; Chao Chen;
879	Trivial or Impossible — Dichotomous Data Difficulty Masks Model Differences (on ImageNet and Beyond) Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: All CNNs make similar decisions since data difficulty is dichotomous.	Kristof Meding; Luca M. Schulze Buschoff; Robert Geirhos; Felix A. Wichmann;
880	Trust Region Policy Optimisation in Multi-Agent Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper introduces the first trust region method for multi-agent reinforcement learning that enjoys theoretically-justified monotonic improvement guarantee and demonstrates the state-of-the-art performance on Mujoco benchmarks.	Jakub Grudzien Kuba; Ruiqing Chen; Muning Wen; Ying Wen; Fanglei Sun; Jun Wang; Yaodong Yang;
881	Tuformer: Data-Driven Design of Expressive Transformer By Tucker Tensor Representation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose Tuformer, a data-driven design of theoretically guaranteed expressive Transformer with trainable heads, inspired by Tucker tensor representation.	Xiaoyu Liu; Jiahao Su; Furong Huang;
882	T-WaveNet: A Tree-Structured Wavelet Neural Network for Time Series Signal Analysis Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we propose a novel tree-structured wavelet neural network for time series signal analysis, namely T-WaveNet, by taking advantage of an inherent property of various types of signals, known as the dominant frequency range.	Minhao LIU; Ailing Zeng; Qiuxia LAI; Ruiyuan Gao; Min Li; Jing Qin; Qiang Xu;
883	Uncertainty Modeling for Out-of-Distribution Generalization Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We for the first time treat feature statistics as uncertain distributions to improve the model generalization ability.	Xiaotong Li; Yongxing Dai; Yixiao Ge; Jun Liu; Ying Shan; LINGYU DUAN;
884	Understanding and Improving Graph Injection Attack By Promoting Unnoticeability Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We find GIA can be provably more harmful than GMA while at the price of bringing more damage to the homophily distribution, which makes it can easily defendable, hence we propose a novel adversarial objective to mitigate the issue.	Yongqiang Chen; Han Yang; Yonggang Zhang; MA KAILI; Tongliang Liu; Bo Han; James Cheng;
885	Understanding and Leveraging Overparameterization in Recursive Value Estimation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present an analysis of value estimation under overparameterized linear representations, and develop new algorithmic tools for improving recursive value estimation with deep models based on the new findings.	Chenjun Xiao; Bo Dai; Jincheng Mei; Oscar A Ramirez; Ramki Gummadi; Chris Harris; Dale Schuurmans;
886	Understanding and Preventing Capacity Loss in Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We show that RL agents experience representation collapse in sparse reward environments and propose an auxiliary task that prevents this from happening and outperforms the state of the art on the Atari benchmark.	Clare Lyle; Mark Rowland; Will Dabney;
887	Understanding Approximate and Unrolled Dictionary Learning for Pattern Recovery Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work studies an approximate formulation of dictionary learning based on unrolling and compares it to alternating minimization to find the best trade-off between speed and precision.	Beno�t Mal�zieux; Thomas Moreau; Matthieu Kowalski;
888	Understanding Dimensional Collapse in Contrastive Self-supervised Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We observe and theoretically explain the dimensional collapse in contrastive self-supervised learning. We also propose a novel SSL method that does not rely on a projector.	Li Jing; Pascal Vincent; Yann LeCun; Yuandong Tian;
889	Understanding Domain Randomization for Sim-to-real Transfer Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose theoretical frameworks for sim-to-real transfer and domain randomization, and provide bounds on the sub-optimality gap of the policy returned by domain randomization.	Xiaoyu Chen; Jiachen Hu; Chi Jin; Lihong Li; Liwei Wang;
890	Understanding Intrinsic Robustness Using Label Uncertainty Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We argue that the standard concentration fails to fully characterize the intrinsic robustness of a classification problem, since it ignores data labels which are essential to any classification task.	Xiao Zhang; David Evans;
891	Understanding Latent Correlation-Based Multiview Learning and Self-Supervision: An Identifiability Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work aims to understand latent correlation maximization-based deep multiview learning from a latent component identification viewpoint.	Qi Lyu; Xiao Fu; Weiran Wang; Songtao Lu;
892	Understanding The Role of Self Attention for Efficient Speech Recognition Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We analyze the role of self attention in Transformer-based speech recognition and present a practical technique to design a model that accelerates the inference and improve the performance.	Kyuhong Shim; Jungwook Choi; Wonyong Sung;
893	Understanding The Variance Collapse of SVGD in High Dimensions Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Qualitative and quantitative analysis of the variance collapse phenomenon of SVGD in high dimensions.	Jimmy Ba; Murat A Erdogdu; Marzyeh Ghassemi; Shengyang Sun; Taiji Suzuki; Denny Wu; Tianzong Zhang;
894	Unified Visual Transformer Compression Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper proposes a unified ViT compression framework that seamlessly assembles three effective techniques: pruning, layer skipping, and knowledge distillation, which outperforms existing competitors.	Shixing Yu; Tianlong Chen; Jiayi Shen; Huan Yuan; Jianchao Tan; Sen Yang; Ji Liu; Zhangyang Wang;
895	UniFormer: Unified Transformer for Efficient Spatial-Temporal Representation Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a novel Unified transFormer (UniFormer) which seamlessly integrates merits of 3D convolution and spatial-temporal self-attention in a concise transformer format, and achieves new state-of-the-art performances on Something-Something.	Kunchang Li; Yali Wang; Gao Peng; Guanglu Song; Yu Liu; Hongsheng Li; Yu Qiao;
896	Unifying Likelihood-free Inference with Black-box Optimization and Beyond Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a framework to unify likelihood-free inference and black-box sequence design and further propose novel sequence design algorithms based on the framework.	Dinghuai Zhang; Jie Fu; Yoshua Bengio; Aaron Courville;
897	Universal Approximation Under Constraints Is Possible with Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We provide the first universal approximation theorem with exact non-convex constraint satisfaction, and we introduce probabilistic transformer networks to do so.	Anastasis Kratsios; Behnoosh Zamanlooy; Tianlin Liu; Ivan Dokmanic;
898	Universalizing Weak Supervision Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We extend weak supervision frameworks to new settings ? rankings, regression, Riemannian spaces, and more ? with a universal algorithm with theoretical guarantees.	Changho Shin; Winfred Li; Harit Vishwakarma; Nicholas Carl Roberts; Frederic Sala;
899	Unraveling Model-Agnostic Meta-Learning Via The Adaptation Learning Rate Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Theoretical analysis of Model-Agnostic Meta-Learning (MAML) through the inner loop (adaptation) learning rate.	Yingtian Zou; Fusheng Liu; Qianxiao Li;
900	Unsupervised Discovery of Object Radiance Fields Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Inferring object-centric factorized 3D scene representations from a single image, learned without 3D geometry or segmentation supervision.	Hong-Xing Yu; Leonidas Guibas; Jiajun Wu;
901	Unsupervised Disentanglement with Tensor Product Representations on The Torus Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Decomposition of a latent space on a torus leads to a disentangled representation	Michael Rotman; Amit Dekel; Shir Gur; Yaron Oz; Lior Wolf;
902	Unsupervised Federated Learning Is Possible Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Federated learning: no label no cry	Nan Lu; Zhao Wang; Xiaoxiao Li; Gang Niu; Qi Dou; Masashi Sugiyama;
903	Unsupervised Learning of Full-Waveform Inversion: Connecting CNN and Partial Differential Equation in A Loop Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We develop an unsupervised method to solve seismic full-waveform inversion in geophysics by integrating CNN and the governing partial differential equation. We also introduce a new large-scale dataset $\textit{OpenFWI}$, to establish a more challenging benchmark for the community.	Peng Jin; Xitong Zhang; Yinpeng Chen; Sharon X Huang; Zicheng Liu; Youzuo Lin;
904	Unsupervised Semantic Segmentation By Distilling Feature Correspondences Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We use the correlations between self-supervised visual features to perform unsupervised semantic segmentation.	Mark Hamilton; Zhoutong Zhang; Bharath Hariharan; Noah Snavely; William T. Freeman;
905	Using Graph Representation Learning with Schema Encoders to Measure The Severity of Depressive Symptoms Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We encode word embeddings by using graph representation learning method, which schematizes context-level depressive features for depression state prediction.	Simin Hong; Anthony Cohn; David Crossland Hogg;
906	VAE Approximation Error: ELBO and Exponential Families Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: VAEs have an inductive bias towards RBMs and generalized linear models	Alexander Shekhovtsov; Dmitrij Schlesinger; Boris Flach;
907	Value Function Spaces: Skill-Centric State Abstractions for Long-Horizon Reasoning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We introduce value function spaces, a learned representation of state through the values of low-level skills, which capture affordances and ignores distractors to enable long-horizon reasoning and zero-shot generalization.	Dhruv Shah; Alexander T Toshev; Sergey Levine; brian ichter;
908	Value Gradient Weighted Model-Based Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose the Value-gradient weighted Model loss, a method for value-aware model learning in challenging settings, such as small model capacity and the presence of distracting state dimensions.	Claas A Voelcker; Victor Liao; Animesh Garg; Amir-massoud Farahmand;
909	Variational Autoencoders in The Presence of Low-dimensional Data: Landscape and Implicit Bias Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We analyze the landscape and implicit optimization bias of VAEs in the presence of low-dimensional data and the implications thereof.	Frederic Koehler; Viraj Mehta; Chenghui Zhou; Andrej Risteski;
910	Variational Methods for Simulation-based Inference Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We combine likelihood-estimation with variational inference to achieve a scalable approach for simulation-based inference.	Manuel Gl�ckler; Michael Deistler; Jakob H. Macke;
911	Variational Neural Cellular Automata Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose and evaluate the Variational Neural Cellular Automata, a self-organising generative model based on neural cellular automata	Rasmus Berg Palm; Miguel Gonz�lez Duque; Shyam Sudhakaran; Sebastian Risi;
912	Variational Oracle Guiding for Reinforcement Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a variational Bayes framework leveraging oracle (hindsight) information available in training to improve deep reinforcement learning.	Dongqi Han; Tadashi Kozuno; Xufang Luo; Zhao-Yun Chen; Kenji Doya; Yuqing Yang; Dongsheng Li;
913	Variational Predictive Routing with Nested Subjective Timescales Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Variational inference hierarchical model that relies on a change detection mechanism to impose a nested temporal hierarchy on its latent structure.	Alexey Zakharov; Qinghai Guo; Zafeirios Fountas;
914	VAT-Mart: Learning Visual Action Trajectory Proposals for Manipulating 3D ARTiculated Objects Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel interaction-for-perception framework to learn visual actionable representations (i.e. affordance and action trajectory proposals) for robotic manipulation.	Ruihai Wu; Yan Zhao; Kaichun Mo; Zizheng Guo; Yian Wang; Tianhao Wu; Qingnan Fan; Xuelin Chen; Leonidas Guibas; Hao Dong;
915	VC Dimension of Partially Quantized Neural Networks in The Overparametrized Regime Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We apply VC theory to analyze the performance of a neural network in the overparametrized regime and obtain a minimax-optimality result.	Yutong Wang; Clayton Scott;
916	Vector-quantized Image Modeling with Improved VQGAN Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose the ViT-VQGAN and further explore a Vector-quantized Image Modeling (VIM) approach on both generative and discriminative tasks on images.	Jiahui Yu; Xin Li; Jing Yu Koh; Han Zhang; Ruoming Pang; James Qin; Alexander Ku; Yuanzhong Xu; Jason Baldridge; Yonghui Wu;
917	VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Variance regularization prevents collapse in self-supervised representation learning	Adrien Bardes; Jean Ponce; Yann LeCun;
918	ViDT: An Efficient and Effective Fully Transformer-based Object Detector Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We integrate vision and detection transformers to build an efficient and effective fully transformer-based object detector.	Hwanjun Song; Deqing Sun; Sanghyuk Chun; Varun Jampani; Dongyoon Han; Byeongho Heo; Wonjae Kim; Ming-Hsuan Yang;
919	Visual Representation Learning Does Not Generalize Strongly Within The Same Domain Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We study and benchmark the inductive biases for generalization in visual representation learning on systematic out-of-distribution settings.	Lukas Schott; Julius Von K�gelgen; Frederik Tr�uble; Peter Vincent Gehler; Chris Russell; Matthias Bethge; Bernhard Sch�lkopf; Francesco Locatello; Wieland Brendel;
920	Visual Representation Learning Over Latent Domains Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A new setting for learning over data with multiple latent domains is introduced, alongside new methods for it.	Lucas Deecke; Timothy Hospedales; Hakan Bilen;
921	ViTGAN: Training GANs with Vision Transformers Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Recently, Vision Transformers (ViTs) have shown competitive performance on image recognition while requiring less vision-specific inductive biases. In this paper, we investigate if such performance can be extended to image generation.	Kwonjoon Lee; Huiwen Chang; Lu Jiang; Han Zhang; Zhuowen Tu; Ce Liu;
922	VOS: Learning What You Don’t Know By Virtual Outlier Synthesis Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we present VOS, a novel framework for OOD detection by adaptively synthesizing virtual outliers that can meaningfully regularize the model’s decision boundary during training.	Xuefeng Du; Zhaoning Wang; Mu Cai; Yixuan Li;
923	W-CTC: A Connectionist Temporal Classification Loss with Wild Cards Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes wild-card CTC to solve the problem that the label only matches middle part of the sequence.	Xingyu Cai; Jiahong Yuan; Yuchen Bian; Guangxu Xun; Jiaji Huang; Kenneth Church;
924	WeakM3D: Towards Weakly Supervised Monocular 3D Object Detection Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This paper explores the weakly supervised monocular 3D detection to dispense with the reliance on 3D box labels.	Liang Peng; Senbo Yan; Boxi Wu; Zheng Yang; Xiaofei He; Deng Cai;
925	Weighted Training for Cross-Task Learning Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We introduce a weighted training algorithm for cross-task learning based on minimizing a representation-based task distance between the source and target tasks.	Shuxiao Chen; Koby Crammer; Hangfeng He; Dan Roth; Weijie J Su;
926	What Do We Mean By Generalization in Federated Learning? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a framework for better measuring generalization and heterogeneity in federated learning, apply it for extensive empirical evaluation across six tasks, and make a series of recommendations for future FL works.	Honglin Yuan; Warren Richard Morningstar; Lin Ning; Karan Singhal;
927	What Happens After SGD Reaches Zero Loss? –A Mathematical Framework Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a mathematical framework to study the implicit bias of SGD after reaching zero loss, based on which we prove label noise can help SGD escape the kernel regime and achieve optimal sample complexity for overparametrized linear model.	Zhiyuan Li; Tianhao Wang; Sanjeev Arora;
928	What Makes Better Augmentation Strategies? Augment Difficult But Not Too Different Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: Effective learning-based augmentation in NLP tasks by constructing difficult but not too different samples	Jaehyung Kim; Dongyeop Kang; Sungsoo Ahn; Jinwoo Shin;
929	When Can We Learn General-Sum Markov Games with A Large Number of Players Sample-Efficiently? Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We present new algorithms for several learning goals in multi-player general-sum Markov games, with mild PAC sample complexity in terms of the number of players.	Ziang Song; Song Mei; Yu Bai;
930	When Should Agents Explore? Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A fresh look at the question of when* to switch into exploration mode, and for how long.*	Miruna Pislar; David Szepesvari; Georg Ostrovski; Diana L Borsa; Tom Schaul;
931	When Vision Transformers Outperform ResNets Without Pre-training or Strong Data Augmentations Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Hence, this paper investigates ViTs and MLP-Mixers from the lens of loss geometry, intending to improve the models’ data efficiency at training and generalization at inference.	Xiangning Chen; Cho-Jui Hsieh; Boqing Gong;
932	When, Why, and Which Pretrained GANs Are Useful? Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: This work aims to dissect the process of GAN finetuning.	Timofey Grigoryev; Andrey Voynov; Artem Babenko;
933	Which Shortcut Cues Will DNNs Choose? A Study from The Parameter-Space Perspective Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: When given equally likely shortcuts in data, which shortcut cue will a DNN choose, and why?	Luca Scimeca; Seong Joon Oh; Sanghyuk Chun; Michael Poli; Sangdoo Yun;
934	Who Is The Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RL Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We theoretically characterize the essence of evasion attacks in RL, and propose a novel attack algorithm for RL agents, which achieves state-of-the-art performance on both attacking and robustifying RL agents in many Atari and MuJoCo tasks.	Yanchao Sun; Ruijie Zheng; Yongyuan Liang; Furong Huang;
935	Who Is Your Right Mixup Partner in Positive and Unlabeled Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a novel PU learning method named P3Mix which simultaneously benefits from instance augmentation and supervision correction with a heuristic mixup technique.	Changchun Li; Ximing Li; Lei Feng; Jihong Ouyang;
936	Why Propagate Alone? Parallel Use of Labels and Features on Graphs Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We analyze a powerful label trick from a theoretical perspective and reduce it to an interpretable form that inspires broader application scenarios.	Yangkun Wang; Jiarui Jin; Weinan Zhang; Yang Yongyi; Jiuhai Chen; Quan Gan; Yong Yu; Zheng Zhang; Zengfeng Huang; David Wipf;
937	Wiring Up Vision: Minimizing Supervised Synaptic Updates Needed to Produce A Primate Ventral Stream Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We develop biologically-motivated initialization and training procedures to train models with 200x fewer synaptic updates (epochs x labeled images x weights) while maintaining 80% of brain predictivity on a set of neural and behavioral benchmarks.	Franziska Geiger; Martin Schrimpf; Tiago Marques; James J. DiCarlo;
938	Wisdom of Committees: An Overlooked Approach To Faster and More Accurate Models Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: A simple ensemble or cascade of off-the-shelf pre-trained models can match or exceed the accuracy of SOTA models while being drastically more efficient.	Xiaofang Wang; Dan Kondratyuk; Eric Christiansen; Kris M. Kitani; Yair Movshovitz-Attias; Elad Eban;
939	Wish You Were Here: Hindsight Goal Selection for Long-horizon Dexterous Manipulation Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: In this work, we extend hindsight relabelling mechanisms to guide exploration along task-specific distributions implied by a small set of successful demonstrations.	Todor Davchev; Oleg Olegovich Sushkov; Jean-Baptiste Regli; Stefan Schaal; Yusuf Aytar; Markus Wulfmeier; Jon Scholz;
940	X-model: Improving Data Efficiency in Deep Learning with A Minimax Model Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: This paper proposes a novel X-Model for improving data efficiency in deep learning with a minimax model.	Ximei Wang; Xinyang Chen; Jianmin Wang; Mingsheng Long;
941	You Are AllSet: A Multiset Function Framework for Hypergraph Neural Networks Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: We propose a multiset function framework for hypergraph neural networks.	Eli Chien; Chao Pan; Jianhao Peng; Olgica Milenkovic;
942	You Mostly Walk Alone: Analyzing Feature Attribution in Trajectory Prediction Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We propose a Shapley value-based method for attributing trajectory prediction performance to different input features and show on common benchmark datasets that existing models do not use interaction information, contrary to their claims.	Osama Makansi; Julius Von K�gelgen; Francesco Locatello; Peter Vincent Gehler; Dominik Janzing; Thomas Brox; Bernhard Sch�lkopf;
943	Zero-CL: Instance and Feature Decorrelation for Negative-free Symmetric Contrastive Learning Related Papers Related Patents Related Grants Related Orgs Related Experts View Highlight: We develop two contrastive learning methods to prevent collapses in symmetric architecture without negative pairs.	Shaofeng Zhang; Feng Zhu; Junchi Yan; Rui Zhao; Xiaokang Yang;
944	Zero Pixel Directional Boundary By Vector Transform Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: In this paper, we re-interpret boundaries as 1-D surfaces and formulate a one-to-one vector transform function that allows for training of boundary prediction completely avoiding the class imbalance issue.	Edoardo Mello Rella; Ajad Chhatkuli; Yun Liu; Ender Konukoglu; Luc Van Gool;
945	Zero-Shot Self-Supervised Learning for MRI Reconstruction Related Papers Related Patents Related Grants Related Orgs Related Experts Related Code View Highlight: Zero-shot self-supervised learning to perform robust subject-specific MRI reconstruction	Burhaneddin Yaman; Seyed Amir Hossein Hosseini; Mehmet Akcakaya;