Dylan J Foster

Correcting the Mythos of KL-Regularization: Direct Alignment without Overoptimization via Chi-Squared Preference Optimization ICLR 2025 Exploratory Preference Optimization: Harnessing Implicit Q*-Approximation for Sample-Efficient RLHF ICLR 2025 Is a Good Foundation Necessary for Efficient Reinforcement Learning? The Computational Role of the Base Model in Exploration COLT 2025 Is Best-of-N the Best of Them? Coverage, Scaling, and Optimality in Inference-Time Alignment ICML 2025 Self-Improvement in Language Models: The Sharpening Mechanism ICLR 2025 Assouad, Fano, and Le Cam with Interaction: A Unifying Lower Bound Framework and Characterization for Bandit Learnability NeurIPS 2024 Butterfly Effects of SGD Noise: Error Amplification in Behavior Cloning and Autoregression ICLR 2024 Can large language models explore in-context? NeurIPS 2024 Harnessing Density Ratios for Online Reinforcement Learning ICLR 2024 Is Behavior Cloning All You Need? Understanding Horizon in Imitation Learning NeurIPS 2024 Online Estimation via Offline Estimation: An Information-Theoretic Framework NeurIPS 2024 Reinforcement Learning Under Latent Dynamics: Toward Statistical and Algorithmic Modularity NeurIPS 2024 Rich-Observation Reinforcement Learning with Continuous Latent Dynamics ICML 2024 Scalable Online Exploration via Coverability ICML 2024 The Power of Resets in Online Reinforcement Learning NeurIPS 2024 Contextual Bandits with Packing and Covering Constraints: A Modular Lagrangian Approach via Regression COLT 2023 Efficient Model-Free Exploration in Low-Rank MDPs NeurIPS 2023 Hardness of Independent Learning and Sparse Equilibrium Computation in Markov Games ICML 2023 Model-Free Reinforcement Learning with the Decision-Estimation Coefficient NeurIPS 2023 Representation Learning with Multi-Step Inverse Kinematics: An Efficient and Optimal Approach to Rich-Observation RL ICML 2023 The Role of Coverage in Online Reinforcement Learning ICLR 2023 Contextual Bandits with Large Action Spaces: Made Practical ICML 2022 Interaction-Grounded Learning with Action-Inclusive Feedback NeurIPS 2022 Offline Reinforcement Learning: Fundamental Barriers for Value Function Approximation COLT 2022 On the Complexity of Adversarial Decision Making NeurIPS 2022 Sample-Efficient Reinforcement Learning in the Presence of Exogenous Information COLT 2022 Understanding the Eluder Dimension NeurIPS 2022 Efficient First-Order Contextual Bandits: Prediction, Allocation, and Triangular Discrimination NeurIPS 2021 Adapting to Misspecification in Contextual Bandits NeurIPS 2020 Independent Policy Gradient Methods for Competitive Reinforcement Learning NeurIPS 2020 Learning the Linear Quadratic Regulator from Nonlinear Observations NeurIPS 2020 Hypothesis Set Stability and Generalization NeurIPS 2019 Model Selection for Contextual Bandits NeurIPS 2019 Contextual bandits with surrogate losses: Margin bounds and efficient algorithms NeurIPS 2018 Uniform Convergence of Gradients for Non-Convex Learning and Optimization NeurIPS 2018 Parameter-Free Online Learning via Model Selection NeurIPS 2017 Spectrally-normalized margin bounds for neural networks NeurIPS 2017 Learning in Games: Robustness of Fast Convergence NeurIPS 2016 Adaptive Online Learning NeurIPS 2015