Ziyi Liu

Generalize and Guide: Decomposing Rewards for Few-Shot Inverse Reinforcement Learning

Learning Generalizable and Well-Shaped Reward Functions from Too Few Demonstrations