Knowledge Commons of Institute of Automation,CAS
Wd3: Taming the estimation bias in deep reinforcement learning | |
He Q(何强)1,2![]() ![]() | |
2020-12 | |
会议名称 | 2020 IEEE 32nd International Conference on Tools with Artificial Intelligence (ICTAI) |
会议日期 | 2020-12 |
会议地点 | Baltimore, MD, USA |
摘要 | The overestimation phenomenon caused by function approximation is a well-known issue in value-based reinforcement learning algorithms such as deep Q-networks and DDPG, which could lead to suboptimal policies. To address this issue, TD3 takes the minimum value between a pair of critics, which introduces underestimation bias. By unifying these two opposites, we propose a novel Weighted Delayed Deep Deterministic Policy Gradient algorithm, which can reduce the estimation error and further improve the performance by weighting a pair of critics. We compare the learning process of value function between DDPG, TD3, and our proposed algorithm, which verifies that our algorithm could indeed eliminate the estimation error of value function. We evaluate our algorithm in the OpenAI Gym continuous control tasks, outperforming the state-of-the-art algorithms on every environment tested. |
关键词 | deep reinforcement learning estimation bias neural networks |
学科门类 | 工学 ; 工学::控制科学与工程 ; 工学::计算机科学与技术(可授工学、理学学位) |
DOI | 10.1109/ICTAI50040.2020.00068 |
URL | 查看原文 |
收录类别 | EI |
语种 | 英语 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/48893 |
专题 | 多模态人工智能系统全国重点实验室_脑机融合与认知评估 |
通讯作者 | Hou XW(侯新文) |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences 2.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 | He Q,Hou XW. Wd3: Taming the estimation bias in deep reinforcement learning[C],2020. |
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 | ||
WD3_Taming_the_Estim(2006KB) | 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 |
[He Q(何强)]的文章 |
[Hou XW(侯新文)]的文章 |
百度学术 |
百度学术中相似的文章 |
[He Q(何强)]的文章 |
[Hou XW(侯新文)]的文章 |
必应学术 |
必应学术中相似的文章 |
[He Q(何强)]的文章 |
[Hou XW(侯新文)]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