Adaptive play of Poker by Graham Kendall.pdf

Adaptive play of Poker by Graham Kendall.pdf

ID:33753957

大小:375.33 KB

页数:15页

时间:2019-02-28

Adaptive play of Poker by Graham Kendall.pdf_第1页
Adaptive play of Poker by Graham Kendall.pdf_第2页
Adaptive play of Poker by Graham Kendall.pdf_第3页
Adaptive play of Poker by Graham Kendall.pdf_第4页
Adaptive play of Poker by Graham Kendall.pdf_第5页
资源描述:

《Adaptive play of Poker by Graham Kendall.pdf》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、DEPOSITBONUSESTOGETSTARTEDFullTiltPoker(100%DepositBonusupto$600)PlayOnlinePokerCakePoker(110%DepositBonusupto$600)WilliamHill(100%DepositBonus&$120,000RakeRace)PrestigeCasino($1,500inFREESpinsNoDepositNeeded)PokerStrategy($50usdStartingCapitalFREENoDepositNeeded)WilliamHillBingo($40usdFRE

2、EToStartPlayingNoDepositRequired)AnInvestigationofanAdaptivePokerPlayerGrahamKendallandMarkWilldigTheUniversityofNottingham,SchoolofComputerScience&IT,JubileeCampus,WollatonRoad,NottinghamNG81BB,UKgxk@cs.nott.ac.uk,Mark.Willdig@siemenscomms.co.ukAbstract:Otherworkhasshownthatadaptivelearni

3、ngcanbehighlysuccessfulindevelopingprogramswhichareabletoplaygamesatalevelsimilartohumanplayersand,insomecases,exceedtheabilityofavastmajorityofhumanplayers.Thisstudyusespokertoinvestigatehowadaptationcanbeusedingamesofimperfectinformation.Aninternallearningvalueismanipulatedwhichallowsapo

4、kerplayingagenttodevelopitsplayingstrategyovertime.Theresultssuggestthattheagentisabletolearnhowtoplaypoker,initiallylosing,beforewinningastheplayersstrategybecomesmoredeveloped.Theevolvedplayerperformswellagainstopponentswithdifferentplayingstyles.Somelimitationsofpreviousworkareovercome,

5、suchasdealrotationtoremovethebiasintroducedbyoneplayeralwaysbeingthelasttoact.Thisworkprovidesencouragementthatthisisanareaworthexploringmorefullyinourfuturework.1.IntroductionGameplayinghasalongresearchhistory.ChesshasreceivedparticularinterestculminatinginDeepBluebeatingKasparovin1997,al

6、beitwithspecializedhardware(Hamilton,1997)andbruteforcesearch.However,althougharguably,beinga‘solvedgame’chessstillreceivesinterestasresearchersturntoadaptivelearningtechniqueswhichallowcomputersto‘learn’toplaychess,ratherthanbeing‘told’howitshouldplay(Kendall,2001).Adaptivelearningwasbein

7、gusedforcheckersasfarbackasthe1950’swithSamuel’sseminalwork(1959,re-producedinSamuel,2000).CheckersresearchwouldleadtoJonathanSchaefferdevelopingChinook,whichclaimedtheworldtitlein1994(Schaeffer,1996).LikeDeepBlue,itisarguableifChinookusedAItechniques.Chinookh

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。