Learning Theory

TensorPlan and the Few Actions Lower Bound for Planning in MDPs under Linear Realizability of Optimal Value Functions ALT 2022 The CLRS Algorithmic Reasoning Benchmark ICML 2022 The Complexity of Learning Approval-Based Multiwinner Voting Rules AAAI 2022 The Curse of Passive Data Collection in Batch Reinforcement Learning AISTATS 2022 The Effect of Manifold Entanglement and Intrinsic Dimensionality on Learning AAAI 2022 The Effects of Corpus Choice and Morphosyntax on Multilingual Space Induction EMNLP 2022 The Fragility of Multi-Treebank Parsing Evaluation COLING 2022 The Franz-Parisi Criterion and Computational Trade-offs in High Dimensional Statistics NeurIPS 2022 The Implicit Bias of Benign Overfitting COLT 2022 The Interplay Between Implicit Bias and Benign Overfitting in Two-Layer Linear Networks JMLR 2022 The merged-staircase property: a necessary and nearly sufficient condition for SGD learning of sparse functions on two-layer neural networks COLT 2022 The Neural Race Reduction: Dynamics of Abstraction in Gated Networks ICML 2022 Theoretical analysis of deep neural networks for temporally dependent observations NeurIPS 2022 The Paradox of the Compositionality of Natural Language: A Neural Machine Translation Case Study ACL 2022 The Pareto Frontier of Instance-Dependent Guarantees in Multi-Player Multi-Armed Bandits with no Communication COLT 2022 The Power and Limitation of Pretraining-Finetuning for Linear Regression under Covariate Shift NeurIPS 2022 The Power of Exploiter: Provable Multi-Agent RL in Large State Spaces ICML 2022 The Primacy Bias in Deep Reinforcement Learning ICML 2022 The Privacy Onion Effect: Memorization is Relative NeurIPS 2022 The “Problem” of Human Label Variation: On Ground Truth in Data, Modeling and Evaluation EMNLP 2022 The Role of Interactivity in Structured Estimation COLT 2022 The role of optimization geometry in single neuron learning AISTATS 2022 The Sample Complexity of One-Hidden-Layer Neural Networks NeurIPS 2022 The Teaching Dimension of Regularized Kernel Learners ICML 2022 The Three Stages of Learning Dynamics in High-dimensional Kernel Methods ICLR 2022