Reinforcement Learning Eligibility Trace强化学.pptVIP

下载本文档

131
0
约8.34千字
约 71页
2017-08-16 发布于重庆
举报
版权申诉

Reinforcement Learning Eligibility Trace强化学.ppt

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Reinforcement Learning Eligibility Trace强化学

Reinforcement LearningEligibility Traces 主講人：虞台文大同大學資工所智慧型多媒體研究室 Content n-step TD prediction Forward View of TD(?) Backward View of TD(?) Equivalence of the Forward and Backward Views Sarsa(?) Q(?) Eligibility Traces for Actor-Critic Methods Replacing Traces Implementation Issues Reinforcement LearningEligibility Traces n-Step TD Prediction 大同大學資工所智慧型多媒體研究室 Elementary Methods Monte Carlo vs. TD(0) Monte Carlo observe reward for all steps in an episode TD(0) observed one step only n-Step TD Prediction n-Step TD Prediction Backups n-Step TD Backup Error Reduction Property Example (Random Walk) Example (19-state Random Walk) Exercise (Random Walk) Exercise (Random Walk) Reinforcement LearningEligibility Traces The Forward View of TD(?) 大同大學資工所智慧型多媒體研究室 Averaging n-step Returns We are not limited to simply using n-step TD returns For example, we could take average n-step TD returns like: TD(?) ? ?-Return TD(?) is a method for averaging all n-step backups weight by ?n?1 (time since visitation) Called ?-return Backup using ?-return: TD(?) ? ?-Return TD(?) is a method for averaging all n-step backups weight by ?n?1 (time since visitation) Called ?-return Backup using ?-return: Forward View of TD(?) TD(?) on the Random Walk Reinforcement LearningEligibility Traces The Backward View of TD(?) 大同大學資工所智慧型多媒體研究室 Why Backward View? Forward view is acausal Not implementable Backward view is causal Implementable In the offline case, achieving the same result as the forward view Eligibility Traces Each state is associated with an additional memory variable ? eligibility trace, defined by: Eligibility Traces Each state is associated with an additional memory variable ? eligibility trace, defined by: Eligibility Traces Each state is associated with an additional memory variable ? eligibility trace, defined by: Eligibility ? Recency of Visiting At any time, the traces record which states have recently been visited, where “recently is defined in terms of