Deep reinforcement learning with Experience Replay based on SARSA
Zhao,Dongbin(赵冬斌); Wang,Haitao; Shao,Kun; Zhu,Yuanheng
2016-09-12
会议名称Proceedings of IEEE Symposium Series on Computational Intelligence (SSCI 2016) – Symposium on Adaptive Dynamic Programming and Reinforcement Learning
会议日期2016-9
会议地点*
摘要SARSA, as one kind of on-policy reinforcement learning methods, is integrated with deep learning to solve the video games control problems in this paper. We use deep convolutional neural network to estimate the state-action value, and SARSA learning to update it. Besides, experience replay is introduced to make the training process suitable to scalable machine learning problems. In this way, a new deep reinforcement learning method, called deep SARSA is proposed to solve complicated control problems such as imitating human to play video games. From the experiments results, we can conclude that the deep SARSA learning shows better performances in some aspects than deep Q learning.
关键词Deep Learning Reinforcement Learning Experience Replay q Learning Sarsa Learning
文献类型会议论文
条目标识符http://ir.ia.ac.cn/handle/173211/19877
专题复杂系统管理与控制国家重点实验室_深度强化学习
作者单位Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences, Beijing 100190, China
推荐引用方式
GB/T 7714
Zhao,Dongbin,Wang,Haitao,Shao,Kun,et al. Deep reinforcement learning with Experience Replay based on SARSA[C],2016.
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
SSCI16_paper_338.pdf(1288KB)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
[Zhao,Dongbin(赵冬斌)]的文章
[Wang,Haitao]的文章
[Shao,Kun]的文章
百度学术
百度学术中相似的文章
[Zhao,Dongbin(赵冬斌)]的文章
[Wang,Haitao]的文章
[Shao,Kun]的文章
必应学术
必应学术中相似的文章
[Zhao,Dongbin(赵冬斌)]的文章
[Wang,Haitao]的文章
[Shao,Kun]的文章
相关权益政策
暂无数据
收藏/分享
文件名: SSCI16_paper_338.pdf
格式: Adobe PDF
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。