文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > AgentP model Learning classifier system with associative perception
侵权投诉

AgentP model Learning classifier system with associative perception

Abstract. Aliasing environments present the tasks of increased difficulty for Learning Classifier Systems. Aliasing squares look identical for an agent with limited perceptive power, but may demand a completely different optimal strategy. Thus, the presenc

4.Thereisnoreward-valuebasedreinforcement,thedistance-basedlearningcoe cientsareusedforthereinforcementpurposes.

4.1StructuralelementsoftheAgentPsystem

Thebasicpartofasingleclassi erissimilartothecondition→action→expectationstructureasusedbyStolzmannforACS[17].ItissupplementedwithanIDsystem,thatinheriteditsbasicstructurefromWilson’smemoryex-tension[23],butisextendedwithaveri cationmechanismandhasdeterministiccharacter.

Thepartsoftheclassi erthatrepresentanenvironmentalstatehaveanidenti cationmarkIDandaveri cationsignF.TheIDmarkisusedforre neddi erentiationofaliasingstates.Iftheveri cationsignofastateinaclassi erispositive,thestateisconsideredashavinga xedID.Thus,incaseofaliasing,theagentis”sure”aboutitspositionintheenvironment,iftheperceptiveimage,itisguidedby,hasa xedIDinthecorrespondingpartoftheclassi er.Eachclassi erinAgentPisassociatedwithapredictionDthatrepresentstheclassi er’sd-coe cient(Section4.5),relatedtotheexpecteddistancetofoodattimet+

AgentP model Learning classifier system with associative perception

1.

上一页第5页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email