Knowledge Commons of Institute of Automation,CAS
Mixing Update Q-value for Deep Reinforcement Learning | |
Li Zhunan1,2![]() ![]() | |
2019-09 | |
会议名称 | International Joint Conference on Neural Networks (IJCNN) |
页码 | 1-6 |
会议日期 | 2019/7/14-19 |
会议地点 | Budapest, Hungary |
会议录编者/会议主办者 | IEEE |
出版者 | IEEE |
摘要 | The value-based reinforcement learning methods are known to overestimate action values such as deep Q-learning, which could lead to suboptimal policies. This problem also persists in an actor-critic algorithm. In this paper, we propose a novel mechanism to minimize its effects on both the critic and the actor. Our mechanism builds on Double Q-learning, by mixing update action value based on the minimum and maximum between a pair of critics to limit the overestimation. We then propose a specific adaptation to the Twin Delayed Deep Deterministic policy gradient algorithm (TD3) and show that the resulting algorithm not only reduces the observed overestimations, as hypothesized, but that this also leads to much better performance on several tasks. |
学科门类 | 工学 |
DOI | 10.1109/IJCNN.2019.8852397 |
收录类别 | EI |
语种 | 英语 |
七大方向——子方向分类 | 强化与进化学习 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/39160 |
专题 | 复杂系统认知与决策实验室_智能系统与工程 |
通讯作者 | Hou Xinwen |
作者单位 | 1.Center for Research on Intelligent System and Engineering, Institute of Automation, Chinese Academy of Sciences 2.School of Artificial Intelligence, University of Chinese Academy of Sciences |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Li Zhunan,Hou Xinwen. Mixing Update Q-value for Deep Reinforcement Learning[C]//IEEE:IEEE,2019:1-6. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
PID5846947.pdf(468KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
个性服务 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[Li Zhunan]的文章 |
[Hou Xinwen]的文章 |
百度学术 |
百度学术中相似的文章 |
[Li Zhunan]的文章 |
[Hou Xinwen]的文章 |
必应学术 |
必应学术中相似的文章 |
[Li Zhunan]的文章 |
[Hou Xinwen]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论