Statistical Learning

The Onset of Variance-Limited Behavior for Networks in the Lazy and Rich Regimes ICLR 2023 Theoretical Guarantees of Learning Ensembling Strategies with Applications to Time Series Forecasting ICML 2023 The Pick-to-Learn Algorithm: Empowering Compression for Tight Generalization Bounds and Improved Post-training Performance NeurIPS 2023 The Price of Differential Privacy under Continual Observation ICML 2023 The Rashomon Importance Distribution: Getting RID of Unstable, Single Model-based Variable Importance NeurIPS 2023 There’s No Data like Better Data: Using QE Metrics for MT Data Filtering EMNLP 2023 The Sample Complexity of Approximate Rejection Sampling With Applications to Smoothed Online Learning COLT 2023 The Statistical Scope of Multicalibration ICML 2023 The s-value: evaluating stability with respect to distributional shifts NeurIPS 2023 Thompson Sampling for High-Dimensional Sparse Linear Contextual Bandits ICML 2023 Thresholded linear bandits AISTATS 2023 Tight Bounds on the Hardness of Learning Simple Nonparametric Mixtures COLT 2023 Tighter PAC-Bayes Bounds Through Coin-Betting COLT 2023 Tighter PAC-Bayes Generalisation Bounds by Leveraging Example Difficulty AISTATS 2023 Tilted Sparse Additive Models ICML 2023 Time-uniform confidence bands for the CDF under nonstationarity NeurIPS 2023 TopP&R: Robust Support Estimation Approach for Evaluating Fidelity and Diversity in Generative Models NeurIPS 2023 Toward Better PAC-Bayes Bounds for Uniformly Stable Algorithms NeurIPS 2023 Towards Empirical Process Theory for Vector-Valued Functions: Metric Entropy of Smooth Function Classes ALT 2023 Towards Foundation Models for Scientific Machine Learning: Characterizing Scaling and Transfer Behavior NeurIPS 2023 Towards Reliable Item Sampling for Recommendation Evaluation AAAI 2023 Towards Understanding Generalization of Graph Neural Networks ICML 2023 Towards Understanding Generalization of Macro-AUC in Multi-label Learning ICML 2023 Training Normalizing Flows from Dependent Data ICML 2023 Transformer-Based Language Model Surprisal Predicts Human Reading Times Best with About Two Billion Training Tokens EMNLP 2023