DetH*: Approximate Hierarchical Solution of Large Markov Decision Processes
Name
Lozano-Perez_DetH.pdf
Size
296.18 KB
Format
Adobe PDF
Checksum (MD5)
2ed3686e24b365c6958fa72842058e17
Author(s)
Barry, Jennifer
Kaelbling, Leslie P.
Lozano-Perez, Tomas
Date Issued
July 2011
Journal
Proceedings of the 22nd 2011 International Joint Conference on Artificial Intelligence
Publisher
AAAI Press/International Joint Conferences on Artificial Intelligence
Citation
Barry, Jennifer, Leslie Pack Kaelbling, and Tomás Lozano-Pérez. "DetH*: Approximate Hierarchical Solution of Large Markov Decision Processes. In 22nd 2011 International Joint Conference on Artificial Intelligence, IJCAI-11, Barcelona, Catalonia, Spain, 16–22 July 2011. AAAI Press, (2011): p.1928-1935.
Version
Author's final manuscript
Abstract
This paper presents an algorithm for finding approximately optimal policies in very large Markov decision processes by constructing a hierarchical model and then solving it approximately. It exploits factored representations to achieve compactness and efficiency and to discover connectivity properties of the domain. We provide a bound on the quality of the solutions and give asymptotic analysis of the runtimes; in addition we demonstrate performance on a collection of very large domains. Results show that the quality of resulting policies is very good and the total running times, for both creating and solving the hierarchy, are significantly less than for an optimal factored MDP solver.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
http://ijcai.org/papers11/Papers/IJCAI11-323.pdf
