首页> 美国政府科技报告 >I. Criterion Equivalence in Discrete Dynamic Programming. II. Stochastic Games with Perfect Information and Time Average Payoff
【24h】

I. Criterion Equivalence in Discrete Dynamic Programming. II. Stochastic Games with Perfect Information and Time Average Payoff

机译:I.离散动态规划中的判据等价。 II。具有完美信息和时间平均收益的随机游戏

获取原文

摘要

It is shown that for finite state and action space Markovian decision processes, a policy is 1-optimal if and only if it is average overtaking optimal so the two criteria are equivalent. A counterexample to an alleged extension of the Hardy-Littlewood Theorem is given, and the optimality of stationary strategies for stochastic games of perfect information with time average payoffs is established. (Author)

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号