Reinforcement_Learning_An_Introduction

《Reinforcement_Learning_An_Introduction》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、Sutton&BartoBook:ReinforcementLearning:AnIntroductionReinforcementLearning:AnIntroductionRichardS.SuttonandAndrewG.BartoMITPress,Cambridge,MA,1998ABradfordBookEndorsementsCodeSolutionsFiguresErrataCourseSlidesThisintroductorytextbookonreinforcementlearningistar

2、getedtowardengineersandscientistsinartificialintelligence,operationsresearch,neuralnetworks,andcontrolsystems,andwehopeitwillalsobeofinteresttopsychologistsandneuroscientists.Ifyouwouldliketoorderacopyofthebook,orifyouarequalifiedinstructorandwouldliketoseeanex

3、aminationcopy,pleaseseetheMITPresshomepageforthisbook.Oryoumightbeinterestedinthereviewsatamazon.com.ThereisalsoaJapanesetranslationavailable.Thetableofcontentsofthebookisgivenbelow,withassociatedHTML.TheHTMLversionhasanumberofpresentationproblems,anditstextiss

4、lightlydifferentfromtherealbook,butitmaybeusefulforsomepurposes.●PrefacePartI:TheProblem●1Introduction❍1.1ReinforcementLearning❍1.2Examples❍1.3ElementsofReinforcementLearning❍1.4AnExtendedExample:Tic-Tac-Toefile:///C¦/book/the-book.html(1of5)[28/08/138203:12:45

5、ﾕﾈﾍ]Sutton&BartoBook:ReinforcementLearning:AnIntroduction❍1.5Summary❍1.6HistoryofReinforcementLearning❍1.7BibliographicalRemarks●2EvaluativeFeedback❍2.1Ann-armedBanditProblem❍2.2Action-ValueMethods❍2.3SoftmaxActionSelection❍2.4EvaluationversusInstruction❍2.5Inc

6、rementalImplementation❍2.6TrackingaNonstationaryProblem❍2.7OptimisticInitialValues❍2.8ReinforcementComparison❍2.9PursuitMethods❍2.10AssociativeSearch❍2.11Conclusion❍2.12BibliographicalandHistoricalRemarks●3TheReinforcementLearningProblem❍3.1TheAgent-Environment

7、Interface❍3.2GoalsandRewards❍3.3Returns❍3.4AUnifiedNotationforEpisodicandContinualTasks❍3.5TheMarkovProperty❍3.6MarkovDecisionProcesses❍3.7ValueFunctions❍3.8OptimalValueFunctions❍3.9OptimalityandApproximation❍3.10Summary❍3.11BibliographicalandHistoricalRemarksP

8、artII:ElementaryMethods●4DynamicProgramming❍4.1PolicyEvaluation❍4.2PolicyImprovement❍4.3PolicyIteration❍4.4ValueIterationfile:///C¦/book/the-book.html(2of5)[28/08/138203:12: