The Community for Technology Leaders
Web Intelligence and Intelligent Agent Technology, IEEE/WIC/ACM International Conference on (2009)
Milan, Italy
Sept. 15, 2009 to Sept. 18, 2009
ISBN: 978-0-7695-3801-3
pp: 311-314
ABSTRACT
Dec-POMDPs are worst-case intractable. Recent approximation algorithms have achieved positive results, but performance depends upon parameters set in advance, and little is known about how to choose those settings. We provide an information-theoretic measure of agent influence, shown to be (1) a good indicator of algorithm performance, and (2) a clue to setting algorithm parameters, potentially improving runtime and memory requirements dramatically without sacrificing quality.
INDEX TERMS
artificial intelligence, cooperative systems, distributed control, approximation methods
CITATION
Shlomo Zilberstein, Martin Allen, "Agent Influence and Intelligent Approximation in Multiagent Problems", Web Intelligence and Intelligent Agent Technology, IEEE/WIC/ACM International Conference on, vol. 02, no. , pp. 311-314, 2009, doi:10.1109/WI-IAT.2009.171
98 ms
(Ver 3.3 (11022016))