Multi-task Self-Supervised Adaptation for Reinforcement Learning

Page view(s)
19
Checked on Jun 15, 2024
Multi-task Self-Supervised Adaptation for Reinforcement Learning
Title:
Multi-task Self-Supervised Adaptation for Reinforcement Learning
Journal Title:
2022 IEEE 17th Conference on Industrial Electronics and Applications (ICIEA)
Keywords:
Publication Date:
12 January 2023
Citation:
Wu, K., Chen, Z., Wu, M., Xiang, S., Jin, R., Zhang, L., & Li, X. (2022). Multi-task Self-Supervised Adaptation for Reinforcement Learning. 2022 IEEE 17th Conference on Industrial Electronics and Applications (ICIEA). https://doi.org/10.1109/iciea54703.2022.10006291
Abstract:
Policy adaptation remains one of the key challenges for reinforcement learning (RL). Thus, RL agents often fail to generalize to unseen scenarios. In this paper, we propose to improve the generalization of RL algorithms through multi-task self-supervised adaptation (MSSA). The proposed method is a general paradigm that can be implemented on top of any RL algorithm. It better extracts high-level feature representations from augmented observations through incorporating multiple self-supervised learning tasks with complementary objectives. The selected self-supervision tasks include rotation prediction, inverse dynamics prediction and contrastive learning. It then performs control actions based on the extracted features. The proposed MSSA method consistently outperforms all the baseline methods on diverse complex tasks in the DeepMind Control suite benchmark and sets new state-of-the-art results without incurring longer inference time. It is demonstrated that MSSA has superior generalization capability and is robust to environmental changes.
License type:
Publisher Copyright
Funding Info:
This research / project is supported by the A*STAR - Career Development Award
Grant Reference no. : C210112046
Description:
© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
2158-2297
Files uploaded:

File Size Format Action
multi-task-self-supervised-adaptation-for-reinforcement-learning.pdf 828.45 KB PDF Request a copy