非平稳MDP的期望平均准则

THE EXPECTED AVERAGE CRITERION FOR NONSTATIONARY MDP

本文考虑的是非平稳MDP的期望平均准则，在弱遍历条件下，用概率及鞅论的方法证明了。∈（0）－最优马氏策略的存在性，作为特例，较好地解决了Feinberg和Park在1994年提及的开问题．

In this paper, we consider the expected average criterion for nonstationary MDP.By probability and martingale method, we prove the existence of ∈( 0)-optimal Markov policies under weakly ergodic conditions. As a typical example of this paper, we solve the open problem posed by Feinberg and Park again.