Show simple item record

dc.contributor.authorCazenave, Tristan
dc.date.accessioned2017-01-16T10:43:35Z
dc.date.available2017-01-16T10:43:35Z
dc.date.issued2016
dc.identifier.issn0304-3975
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/16163
dc.language.isoenen
dc.subjectComputer Gamesen
dc.subjectMonte Carlo Tree Searchen
dc.subjectReinforcement Learningen
dc.subjectPlayout policyen
dc.subjectMachine learningen
dc.subject.ddc006.3en
dc.titlePlayout Policy Adaptation with Move Featuresen
dc.typeArticle accepté pour publication ou publié
dc.description.abstractenMonte Carlo Tree Search (MCTS) is the state of the art algorithm for General Game Playing (GGP). We propose to learn a playout policy online so as to improve MCTS for GGP. We also propose to learn a policy not only using the moves but also according to the features of the moves. We test the resulting algorithms named Playout Policy Adaptation (PPA) and Playout Policy Adaptation with move Features (PPAF) on Atarigo, Breakthrough, Misere Breakthrough, Domineering, Misere Domineering, Knightthrough, MisereKnightthrough and Nogo. The experiments compare PPA and PPAF to Upper Confidence for Trees (UCT) and to the closely related Move-Average Sampling Technique (MAST) algorithm.en
dc.relation.isversionofjnlnameTheoretical Computer Science
dc.relation.isversionofjnlvol644en
dc.relation.isversionofjnldate2016-06
dc.relation.isversionofjnlpages43-52en
dc.relation.isversionofdoi10.1016/j.tcs.2016.06.024en
dc.relation.isversionofjnlpublisherElsevieren
dc.subject.ddclabelIntelligence artificielleen
dc.relation.forthcomingnonen
dc.relation.forthcomingprintnonen
dc.description.ssrncandidatenonen
dc.description.halcandidateouien
dc.description.readershiprechercheen
dc.description.audienceInternationalen
dc.relation.Isversionofjnlpeerreviewedouien
dc.relation.Isversionofjnlpeerreviewedouien
dc.date.updated2017-01-16T10:37:58Z
hal.person.labIds989
hal.identifierhal-01436210*


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record