基于未来阈值的马尔可夫策略的最优化算法

Algorithm of Markov optimum policy based on future threshold value

在一种新的准则概率阈值准则下讨论马尔可夫决策的最优解的算法问题.在该准则下,采用基于未来阈值的方法,求解马尔可夫最优策略.

The arithmetic problem of Markov optimum solution under a new principle named probability threshold value principle is discussed.With this principle,the Markov optimum policy is solved based on the future threshold value.