1 / 551
文档名称:

MIT Press - Reinforcement Learning - An Introduction(1).pdf

格式:pdf   页数:551
下载后只包含 1 个 PDF 格式的文档,没有任何的图纸或源代码,查看文件列表

如果您已付费下载过本站文档,您可以点这里二次下载

MIT Press - Reinforcement Learning - An Introduction(1).pdf

上传人:bolee65 2014/4/9 文件大小:0 KB

下载得到文件列表

MIT Press - Reinforcement Learning - An Introduction(1).pdf

文档介绍

文档介绍:ReinforcementLearning:,Cambridge,MA,1998ABradfordBookEndorsementsCodeSolutionsFiguresErrataCourseSlidesThisintroductorytextbookonreinforcementlearningistargetedtowardengineersandscientistsinartificialintelligence,operationsresearch,works,andcontrolsystems,,orifyouarequalifiedinstructorandwouldliketoseeanexaminationcopy,pleaseseetheMITPresshomepageforthisbook..,,anditstextisslightlydifferentfromtherealbook,butitmaybeusefulforsomepurposes.●PrefacePartI:TheProblem●1Introduction❍❍❍❍:Tic-Tac-Toe❍❍❍●2EvaluativeFeedback❍-armedBanditProblem❍-ValueMethods❍❍❍❍❍❍❍❍❍❍●3TheReinforcementLearningProblem❍-EnvironmentInterface❍❍❍❍❍❍❍❍❍❍:ElementaryMethods●4DynamicProgramming❍❍❍❍❍❍❍❍❍●5MonteCarloMethods❍❍❍❍-PolicyMonteCarloControl❍5