Multi-policy iteration with a distributed voting

We present a novel simulation-based algorithm, as an extension of the well-known policy iteration algorithm, by combining multi-policy improvement with a distributed simulation-based voting policy evaluation, for approximately solving Markov Decision Processes (MDPs) with infinite horizon discounted reward criterion, and analyze its performance relative to the optimal value. Copyright Springer-Verlag 2004

MoreLess

Year of publication:	2004
Authors:	Chang, Hyeong Soo
Published in:	Computational Statistics. - Springer. - Vol. 60.2004, 2, p. 299-310
Publisher:	Springer
Subject:	Policy iteration \| Distributed algorithm \| Voting \| Markov decision processes

More details

Extent:	text/html
Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10010847739