Qining Zhang

Provable Policy Optimization for Reinforcement Learning from Trajectory Preferences with an Unknown Link Function

Zeroth-Order Policy Gradient for Reinforcement Learning from Human Feedback without Reward Inference