A policy-improvement type algorithm for solving zero-sum two-person stochastic games of perfect information |
| |
| Authors: | TES Raghavan Zamir Syed |
| |
| Institution: | (1) Department of Mathematics, Statistics and Computer Science, University of Illinois at Chicago, e-mail: ter@uic.edu, US;(2) The Hull Group L.L.C, Chicago, IL 60606, e-mail: zsyed@hdc.com, US |
| |
| Abstract: | We give a policy-improvement type algorithm to locate an optimal pure stationary strategy for discounted stochastic games
with perfect information. A graph theoretic motivation for our algorithm is presented as well.
Received: January 1998 / Accepted: May 2002 Published online: February 14, 2003
Key words. stochastic games – MDP – perfect information – policy iteration
Partially Funded by NSF Grant DMS 930-1052 and DMS 970-4951 |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
