简介：Inthispaperwediscussthediscrete,timenon--homogeneousdiscountedMarkoviandecisionprogramming,wherethestatespaceandallactionsetsarecountable.Supposethattheoptimumvaluefunctionisfinite.Wegivethenecessaryandsufficientconditionsfortheexistenceofanoptimalpolicy.Supposethattheabsolutemeanofrewardsisrelativelybounded.Wealsogivethenecessaryandsufficientconditionsfortheexistenceofanoptimalpolicy.
客服QQ：30444492琼网文【2021】1550-113号
增值电信业务经营许可证：琼B2-20210322
出版物经营许可证：新出发龙华出字第(2021)009号
广播电视节目制作经营许可证：(琼)字第00779号
版权所有 ©2002-2024 期刊网(www.qikanchina.com) 琼ICP备2021005105号
