An Algorithm to Identify and Compute Average Optimal Policies in Multichain Markov Decision Processes
An Algorithm to Identify and Compute Average Optimal Policies in Multichain Markov Decision Processes
复制标题
一种识别和计算多链马尔可夫决策过程中平均最优策略的算法
DOI：
10.1287/moor.28.3.553.16388
复制
发表时间：
2003
期刊：
影响因子：
--
通讯作者：
A. Leizarowitz
中科院分区：
文献类型：
--
作者：
A. Leizarowitz
本文研究了离散时间、有限状态的多链MDP，其作用集是紧的。最优准则是长期平均成本。简单的例子说明，最佳的平稳马尔可夫策略并不总是存在的。我们建立了稳定的马尔可夫的e-最优策略的存在，并开发了一个算法，计算这些近似的最优策略。我们建立了一个必要和充分的条件，存在一个最佳的政策，是平稳的马尔可夫，并在这种情况下，这样的最佳政策存在的算法计算它。
This paper concerns discrete-time, finite state multichain MDPs with compact action sets. The optimality criterion is long-run average cost. Simple examples illustrate that optimal stationary Markov policies do not always exist. We establish the existence of e-optimal policies that are stationary Markovian, and develop an algorithm that computes these approximate optimal policies. We establish a necessary and sufficient condition for the existence of an optimal policy that is stationary Markovian, and in case that such an optimal policy exists the algorithm computes it.