Hierarchical Reinforcement Learning With Automatic Sub-Goal Identification
Chenghao Liu; Fei Zhu; Quan Liu; Yuchen Fu
刊名IEEE/CAA Journal of Automatica Sinica
2021
卷号8期号:10页码:1686-1696
关键词Hierarchical control hierarchical reinforcement learning option sparse reward sub-goal
ISSN号2329-9266
DOI10.1109/JAS.2021.1004141
英文摘要In reinforcement learning an agent may explore ineffectively when dealing with sparse reward tasks where finding a reward point is difficult. To solve the problem, we propose an algorithm called hierarchical deep reinforcement learning with automatic sub-goal identification via computer vision (HADS) which takes advantage of hierarchical reinforcement learning to alleviate the sparse reward problem and improve efficiency of exploration by utilizing a sub-goal mechanism. HADS uses a computer vision method to identify sub-goals automatically for hierarchical deep reinforcement learning. Due to the fact that not all sub-goal points are reachable, a mechanism is proposed to remove unreachable sub-goal points so as to further improve the performance of the algorithm. HADS involves contour recognition to identify sub-goals from the state image where some salient states in the state image may be recognized as sub-goals, while those that are not will be removed based on prior knowledge. Our experiments verified the effect of the algorithm.
内容类型期刊论文
源URL[http://ir.ia.ac.cn/handle/173211/45382]  
专题自动化研究所_学术期刊_IEEE/CAA Journal of Automatica Sinica
推荐引用方式
GB/T 7714
Chenghao Liu,Fei Zhu,Quan Liu,et al. Hierarchical Reinforcement Learning With Automatic Sub-Goal Identification[J]. IEEE/CAA Journal of Automatica Sinica,2021,8(10):1686-1696.
APA Chenghao Liu,Fei Zhu,Quan Liu,&Yuchen Fu.(2021).Hierarchical Reinforcement Learning With Automatic Sub-Goal Identification.IEEE/CAA Journal of Automatica Sinica,8(10),1686-1696.
MLA Chenghao Liu,et al."Hierarchical Reinforcement Learning With Automatic Sub-Goal Identification".IEEE/CAA Journal of Automatica Sinica 8.10(2021):1686-1696.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

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。


©版权所有 ©2017 CSpace - Powered by CSpace