马氏决策向量过程模型初步研究

Preliminary Study on Markov Decision-making Vector Processes

在传统马氏单元决策过程（MDP）模型中引入多元行动来确定系统的状态转移概率,通过运用传统MDP的基本理论以及结合多元行动集、决策向量、相合度等新定义,提出了马氏向量决策过程模型.

This paper studies the multivariate actions to define the state-transition probability during the traditional model of MDP.By applying the Markov decision processes theory and the new definition of multivariate actions set,decision-making vector,consistent degree ETC,the new model of Markov decision-making vector processes is introduced.