WEAK CONDITIONS FOR AVERAGE OPTIMALITY IN MARKOV CONTROL PROCESSES

被引：8
作者：
HERNANDEZLERMA, O
LASSERRE, JB
机构：
[1] INST POLITECN NACL,CINVESTAV,DEPT MATH,MEXICO CITY 07000,DF,MEXICO
[2] CNRS,LAAS,F-31077 TOULOUSE,FRANCE
来源：
SYSTEMS & CONTROL LETTERS | 1994年 / 22卷 / 04期
关键词：
(DISCRETE-TIME) MARKOV CONTROL PROCESSES; AVERAGE COST CRITERION;
D O I：
10.1016/0167-6911(94)90060-4
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
In this paper we study the existence of average cost optimal policies for Markov control processes on Borel spaces and with possibly unbounded costs and controls. Under assumptions weaker than those used in the previous literature, we show that the 'optimality inequality' holds everywhere, and that there exists a stationary policy which is optimal whenever the initial state lies in a possibly proper subset of the state space. This is in contrast to previous works where an implicit 'unichain' assumption ensures optimality for all initial states.
引用
收藏
页码：287 / 291
页数：5
相关论文
共 13 条
[1]  
ARAPOSTATHIS A, IN PRESS SIAM J CONT
[2]   LINEAR PROGRAMMING IN A MARKOV DECISION PROBLEM [J].
DENARDO, EV .
MANAGEMENT SCIENCE SERIES A-THEORY, 1970, 16 (05) :281-288
[3]  
Dynkin EB, 1979, CONTROLLED MARKOV PR
[4]  
Hernandez-Lerma O., 1989, ADAPTIVE MARKOV CONT
[5]  
HERNANDEZLERMA O, 1992, KYBERNETIKA, V28, P191
[6]   AVERAGE COST OPTIMAL POLICIES FOR MARKOV CONTROL PROCESSES WITH BOREL STATE-SPACE AND UNBOUNDED COSTS [J].
HERNANDEZLERMA, O ;
LASSERRE, JB .
SYSTEMS & CONTROL LETTERS, 1990, 15 (04) :349-356
[7]   AVERAGE OPTIMALITY IN DYNAMIC-PROGRAMMING ON BOREL SPACES - UNBOUNDED COSTS AND CONTROLS [J].
HERNANDEZLERMA, O .
SYSTEMS & CONTROL LETTERS, 1991, 17 (03) :237-242
[8]  
HERNANDEZLERMA O, 1993, UNPUB AVERAGE OPTIMA
[9]  
MONTESDEOCA R, 1992, UNPUB CONDITIONS AVE
[10]  
RIEDER U, 1978, MANUSCRIPTA MATH, V24, P507
12