mirage   mirage   mirage

AO* revisited

DSpace/Manakin Repository

Show simple item record

dc.creator Bayer-Zubek, Valentina
dc.creator Dietterich, Thomas Glen
dc.date.accessioned 2012-08-31T17:03:29Z
dc.date.available 2012-08-31T17:03:29Z
dc.date.issued 2004-06-30
dc.identifier.uri http://hdl.handle.net/1957/33122
dc.description Preprint submitted to Elsevier Science 30 June 2004. en_US
dc.description.abstract This paper revisits the AO* algorithm introduced by Martelli and Montanari [1] and made popular by Nilsson [2] The paper's main contributions are: (1) proving that the value of a node monotonically increases as the AO* search progresses , (2) proving that selective updates of the value function in the AO* algorithm are correct and providing guidance to researchers interested in the AO* implementation. (1) and (2) are proven under the assumption that the heuristic used by AO* is consistent. The paper also reviews the use of AO* for solving Markov Decision Processes (MDPs) and Partially Observable Markov Decision Processes (POMDPs). en_US
dc.language.iso en_US en_US
dc.subject Heuristic search en_US
dc.subject AO implementation en_US
dc.subject AND / OR graph en_US
dc.subject admissible heuristic en_US
dc.subject consistent heuristic en_US
dc.subject MDP en_US
dc.subject POMDP en_US
dc.title AO* revisited en_US
dc.type Article en_US


This item appears in the following Collection(s)

Show simple item record

Search ScholarsArchive@OSU


Advanced Search

Browse

My Account

Statistics