Policy-labeled Preference Learning: Is Preference Enough for RLHF?
This addresses a key bottleneck in RLHF for aligning AI systems with human goals, though it appears incremental as it builds on existing frameworks like Direct Preference Optimization.
The paper tackles the problem of inaccurate likelihood estimation in Reinforcement Learning from Human Feedback (RLHF) by proposing policy-labeled preference learning (PPL), which models human preferences with regret to incorporate behavior policy information. Experiments in continuous control tasks show significant improvements in offline RLHF performance and effectiveness in online settings.
To design rewards that align with human goals, Reinforcement Learning from Human Feedback (RLHF) has emerged as a prominent technique for learning reward functions from human preferences and optimizing policies via reinforcement learning algorithms. However, existing RLHF methods often misinterpret trajectories as being generated by an optimal policy, causing inaccurate likelihood estimation and suboptimal learning. Inspired by Direct Preference Optimization framework which directly learns optimal policy without explicit reward, we propose policy-labeled preference learning (PPL), to resolve likelihood mismatch issues by modeling human preferences with regret, which reflects behavior policy information. We also provide a contrastive KL regularization, derived from regret-based principles, to enhance RLHF in sequential decision making. Experiments in high-dimensional continuous control tasks demonstrate PPL's significant improvements in offline RLHF performance and its effectiveness in online settings.