首页 | 本学科首页   官方微博 | 高级检索  
     

Bandit过程及其应用
引用本文:王熙逵. Bandit过程及其应用[J]. 经济数学, 2001, 18(4): 39-48
作者姓名:王熙逵
作者单位:Manitoba大学统计系,加拿大
基金项目:the Natural Sciences and Engineering Research Council( NSERC) of Canada
摘    要:本文有两个目的.第一,对Bandit过程这一学科的主要概念及结果作一次系统性的介绍.第二,综述Bandit过程的模型,计算与应用的最新发展.本文刻画了Bandit过程与马氏决策规划的关系.通过考虑理论上或方法论上的局限,实际中或计算上的困难,以及应用中的限制.我们讨论一些重要的争端和公开的问题.

关 键 词:Bandit过程,动态规划,马氏决策过程,最优决策,"  乘胜"  决策,近视决策

BANDIT PROCESSES ANDAPPLICATIONS
Abstract. BANDIT PROCESSES ANDAPPLICATIONS[J]. Mathematics in Economics, 2001, 18(4): 39-48
Authors:Abstract
Abstract:The objective of this article is twofold:(1)to offer a tutorial introduction to some of the key concepts and results in the subject of bandit processes,and(2) to survey some recent developments on the models,computation,and applications of bandit problems.Bandit problems are formulated as a special class of Markov decision processes.Important issues and open problems are discussed based on the consideration of theoretical or methodological limitations,and practical or computational diffculties,and constraints in applications.
Keywords:Bandit processes  dynamic programming  Markov decision processes  optimal strategy  play-the-winner rule  myopic strategy
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号