Lin Yang

Near-Optimal Sample Complexity for Online Constrained MDPs

Deployment Efficient Reward-Free Exploration with Linear Function Approximation

NoWag: A Unified Framework for Shape Preserving Com- pression of Large Language Models

Trajectory Graph Learning: Aligning with Long Trajectories in Reinforcement Learning Without Reward Design

Regret-Optimal List Replicable Bandit Learning: Matching Upper and Lower Bounds

Misspecified $Q$-Learning with Sparse Linear Function Approximation: Tight Bounds on Approximation Error

Breaking the Frozen Subspace: Importance Sampling for Low-Rank Optimization in LLM Pretraining

Transfer Q-Learning with Composite MDP Structures

Hyper: Hyperparameter Robust Efficient Exploration in Reinforcement Learning

Deployment Efficient Reward-Free Exploration with Linear Function Approximation

Low-Switching Primal-Dual Algorithms for Safe Reinforcement Learning

Uniform Last-Iterate Guarantee for Bandits and Reinforcement Learning

Confident Natural Policy Gradient for Local Planning in $q_\pi$-realizable Constrained MDPs

DEXR: A Unified Approach Towards Environment Agnostic Exploration