The generalized Howard method for finding strategies for controlled Markovchains

Authors
Ibragimov, AA
Citation
Aa. Ibragimov, The generalized Howard method for finding strategies for controlled Markovchains, J COMP SYST, 39(2), 2000, pp. 211-215
Citations number
7
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL
ISSN journal
10642307 → ACNP
Volume
39
Issue
2
Year of publication
2000
Pages
211 - 215
Database
ISI
SICI code
1064-2307(200003/04)39:2<211:TGHMFF>2.0.ZU;2-3
Abstract
A problem of finding an optimal strategy for controlled Markov chains with discounting is considered in the case when the discount factor is close to 1. An iteration algorithm-the generalized Howard method-is constructed, whi ch is more feasible for practical implementation than the known methods.