Learning Theory

Generalization Guarantees for Neural Architecture Search with Train-Validation Split ICML 2021 Generalization in Portfolio-Based Algorithm Selection AAAI 2021 Generalization of Model-Agnostic Meta-Learning Algorithms: Recurring and Unseen Tasks NeurIPS 2021 Generalizing Complex Hypotheses on Product Distributions: Auctions, Prophet Inequalities, and Pandora’s Problem COLT 2021 Generic Oracles for Structured Prediction ACL 2021 Global Convergence of Three-layer Neural Networks in the Mean Field Regime ICLR 2021 Global optimality of softmax policy gradient with single hidden layer neural networks in the mean-field regime ICLR 2021 Gradient Descent on Two-layer Nets: Margin Maximization and Simplicity Bias NeurIPS 2021 Gradient Methods Never Overfit On Separable Data JMLR 2021 Gradient Starvation: A Learning Proclivity in Neural Networks NeurIPS 2021 Graph-based semi-supervised learning through the lens of safety UAI 2021 Grounding inductive biases in natural images: invariance stems from variations in data NeurIPS 2021 Hardness of Identity Testing for Restricted Boltzmann Machines and Potts models JMLR 2021 Hessian Eigenspectra of More Realistic Nonlinear Models NeurIPS 2021 High-Confidence Off-Policy (or Counterfactual) Variance Estimation AAAI 2021 How Do Neural Sequence Models Generalize? Local and Global Cues for Out-of-Distribution Prediction EMNLP 2021 How Important is the Train-Validation Split in Meta-Learning? ICML 2021 How Much Over-parameterization Is Sufficient to Learn Deep ReLU Networks? ICLR 2021 How Neural Networks Extrapolate: From Feedforward to Graph Neural Networks ICLR 2021 How Powerful are Performance Predictors in Neural Architecture Search? NeurIPS 2021 How Robust are Model Rankings : A Leaderboard Customization Approach for Equitable Evaluation AAAI 2021 How rotational invariance of common kernels prevents generalization in high dimensions ICML 2021 How Tight Can PAC-Bayes be in the Small Data Regime? NeurIPS 2021 HyDRA: Hypergradient Data Relevance Analysis for Interpreting Deep Neural Networks AAAI 2021 Hypothesis testing with low-degree polynomials in the Morris class of exponential families COLT 2021