Prioritized offline Goal-swapping Experience Replay

Wenyan Yang, Joni Pajarinen, Dingding Cai, Joni-kristian Kamarainen

Keywords: offline reinforcement learning, goal-conditioned reinforcement learning, experience replay

Abstract: In goal-conditioned offline reinforcement learning, an agent learns from previously collected data to go to an arbitrary goal. Since the offline data only contains a finite number of trajectories, a main challenge is how to generate more data. Goal-swapping generates additional data by switching trajectory goals but while doing so produces a large number of invalid trajectories. To address this issue, we propose prioritized goal-swapping experience replay (PGSER). PGSER uses a pre-trained Q function to assign higher priority weights to goal swapped transitions that allow reaching the goal. In experiments, PGSER significantly improves over baselines in a wide range of benchmark tasks, including challenging previously unsuccessful dexterous in-hand manipulation tasks.

Track: Technical Paper

Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.