基于马氏决策向量过程模型的有限阶段期望总报酬准则及其最优方程

The Finite Stage of Expected Totall Reward Model and Optimality Equation of Markov in Decision-making Vector Processes

在马氏决策向量过程模型的理论基础上,结合决策向量和相合度等新定义,进一步提出有限阶段期望总报酬准则和最优方程,并证明最优方程的解的存在性.

By applying Markov decision - making vector processes theory and the new definition of decision - making vector, consistency degree, ETC. This paper will study the finite stage of expected totall reward model and optimality equation in Markov decision - making vector processes. Finally we proved the existence of solutions in the optimality equation.