Game-Theoretic Robust Reinforcement Learning Handles Temporally-Coupled Perturbations
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911852010668032 |
|---|---|
| author | Liang, Yongyuan Sun, Yanchao Zheng, Ruijie Liu, Xiangyu Eysenbach, Benjamin Sandholm, Tuomas Huang, Furong McAleer, Stephen |
| author_facet | Liang, Yongyuan Sun, Yanchao Zheng, Ruijie Liu, Xiangyu Eysenbach, Benjamin Sandholm, Tuomas Huang, Furong McAleer, Stephen |
| contents | Deploying reinforcement learning (RL) systems requires robustness to uncertainty and model misspecification, yet prior robust RL methods typically only study noise introduced independently across time. However, practical sources of uncertainty are usually coupled across time. We formally introduce temporally-coupled perturbations, presenting a novel challenge for existing robust RL methods. To tackle this challenge, we propose GRAD, a novel game-theoretic approach that treats the temporally-coupled robust RL problem as a partially observable two-player zero-sum game. By finding an approximate equilibrium within this game, GRAD optimizes for general robustness against temporally-coupled perturbations. Experiments on continuous control tasks demonstrate that, compared with prior methods, our approach achieves a higher degree of robustness to various types of attacks on different attack domains, both in settings with temporally-coupled perturbations and decoupled perturbations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2307_12062 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Game-Theoretic Robust Reinforcement Learning Handles Temporally-Coupled Perturbations Liang, Yongyuan Sun, Yanchao Zheng, Ruijie Liu, Xiangyu Eysenbach, Benjamin Sandholm, Tuomas Huang, Furong McAleer, Stephen Machine Learning Artificial Intelligence Deploying reinforcement learning (RL) systems requires robustness to uncertainty and model misspecification, yet prior robust RL methods typically only study noise introduced independently across time. However, practical sources of uncertainty are usually coupled across time. We formally introduce temporally-coupled perturbations, presenting a novel challenge for existing robust RL methods. To tackle this challenge, we propose GRAD, a novel game-theoretic approach that treats the temporally-coupled robust RL problem as a partially observable two-player zero-sum game. By finding an approximate equilibrium within this game, GRAD optimizes for general robustness against temporally-coupled perturbations. Experiments on continuous control tasks demonstrate that, compared with prior methods, our approach achieves a higher degree of robustness to various types of attacks on different attack domains, both in settings with temporally-coupled perturbations and decoupled perturbations. |
| title | Game-Theoretic Robust Reinforcement Learning Handles Temporally-Coupled Perturbations |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2307.12062 |