English

Online Policy Learning from Offline Preferences

Machine Learning 2024-03-18 v1

Abstract

In preference-based reinforcement learning (PbRL), a reward function is learned from a type of human feedback called preference. To expedite preference collection, recent works have leveraged \emph{offline preferences}, which are preferences collected for some offline data. In this scenario, the learned reward function is fitted on the offline data. If a learning agent exhibits behaviors that do not overlap with the offline data, the learned reward function may encounter generalizability issues. To address this problem, the present study introduces a framework that consolidates offline preferences and \emph{virtual preferences} for PbRL, which are comparisons between the agent's behaviors and the offline data. Critically, the reward function can track the agent's behaviors using the virtual preferences, thereby offering well-aligned guidance to the agent. Through experiments on continuous control tasks, this study demonstrates the effectiveness of incorporating the virtual preferences in PbRL.

Keywords

Cite

@article{arxiv.2403.10160,
  title  = {Online Policy Learning from Offline Preferences},
  author = {Guoxi Zhang and Han Bao and Hisashi Kashima},
  journal= {arXiv preprint arXiv:2403.10160},
  year   = {2024}
}
R2 v1 2026-06-28T15:21:31.686Z