Revisiting Discrete Soft Actor-Critic
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2022
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916488746631168 |
|---|---|
| author | Zhou, Haibin Wei, Tong Lin, Zichuan li, junyou Xing, Junliang Shi, Yuanchun Shen, Li Yu, Chao Ye, Deheng |
| author_facet | Zhou, Haibin Wei, Tong Lin, Zichuan li, junyou Xing, Junliang Shi, Yuanchun Shen, Li Yu, Chao Ye, Deheng |
| contents | We study the adaption of Soft Actor-Critic (SAC), which is considered as a state-of-the-art reinforcement learning (RL) algorithm, from continuous action space to discrete action space. We revisit vanilla discrete SAC and provide an in-depth understanding of its Q value underestimation and performance instability issues when applied to discrete settings. We thereby propose Stable Discrete SAC (SDSAC), an algorithm that leverages entropy-penalty and double average Q-learning with Q-clip to address these issues. Extensive experiments on typical benchmarks with discrete action space, including Atari games and a large-scale MOBA game, show the efficacy of our proposed method. Our code is at: https://github.com/coldsummerday/SD-SAC.git. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2209_10081 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Revisiting Discrete Soft Actor-Critic Zhou, Haibin Wei, Tong Lin, Zichuan li, junyou Xing, Junliang Shi, Yuanchun Shen, Li Yu, Chao Ye, Deheng Machine Learning Artificial Intelligence We study the adaption of Soft Actor-Critic (SAC), which is considered as a state-of-the-art reinforcement learning (RL) algorithm, from continuous action space to discrete action space. We revisit vanilla discrete SAC and provide an in-depth understanding of its Q value underestimation and performance instability issues when applied to discrete settings. We thereby propose Stable Discrete SAC (SDSAC), an algorithm that leverages entropy-penalty and double average Q-learning with Q-clip to address these issues. Extensive experiments on typical benchmarks with discrete action space, including Atari games and a large-scale MOBA game, show the efficacy of our proposed method. Our code is at: https://github.com/coldsummerday/SD-SAC.git. |
| title | Revisiting Discrete Soft Actor-Critic |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2209.10081 |