Seoung Bum Kim

Policy Likelihood-based Query Sampling and Critic-Exploited Reset for Efficient Preference-based Reinforcement Learning

Multi-Expert Distributionally Robust Optimization for Out-of-Distribution Generalization