Offline goal-conditioned reinforcement learning (GCRL) can be challenging due to overfitting to the given dataset. To generalize agents' skills outside the given dataset, we propose a goal-swapping procedure that generates additional trajectories. To alleviate the problem of noise and extrapolation errors, we present a general offline reinforcement learning method called deterministic Q-advantage policy gradient (DQAPG). In the experiments, DQAPG outperforms state-of-the-art goal-conditioned offline RL methods in a wide range of benchmark tasks, and goal-swapping further improves the test results. It is noteworthy, that the proposed method obtains good performance on the challenging dexterous in-hand manipulation tasks for which the prior methods failed.
@article{arxiv.2302.08865,
title = {Swapped goal-conditioned offline reinforcement learning},
author = {Wenyan Yang and Huiling Wang and Dingding Cai and Joni Pajarinen and Joni-Kristen Kämäräinen},
journal= {arXiv preprint arXiv:2302.08865},
year = {2023}
}
Comments
arXiv admin note: text overlap with arXiv:2302.07741