Ke Jiang

Koopman-Assisted Trajectory Synthesis: A Data Augmentation Framework for Offline Imitation Learning

Learning to Distinguish: Behavior Gap Optimization for Goal-Conditioned Policy Learning

Decoupling Policy Improvement and Conservatism in Offline Reinforcement Learning

Towards Reliable Offline Reinforcement Learning via Lyapunov Uncertainty Control

Outcome-Driven Action Flexibility for Robust Offline Reinforcement Learning