摘要
本文定义了折扣因子可以取不同值的折扣多目标马氏决策规划(DMOMDP),讨论了它的马氏策略(П_m^d)与平稳策略(П_s^d)的优势及局限性。
In this paper, Discounted Multi-Objective Markov Decision Programming (DMOMDP) is defined, and superiorities and limitations of the Markov strategy (Ⅱ_m^d) and stationary strategy (Ⅱ_m^d) are discussed.
出处
《桂林电子工业学院学报》
1989年第1期18-23,共6页
Journal of Guilin Institute of Electronic Technology
关键词
多目标规划
马氏策略
平稳策略
multi-objective programming
Markov strategy
stationary strategy
optimal strategy
