Show simple item record

dc.contributor.authorAnthony, Tom
dc.contributor.authorPolani, D.
dc.contributor.authorNehaniv, C.L.
dc.date.accessioned2015-02-17T15:48:14Z
dc.date.available2015-02-17T15:48:14Z
dc.date.issued2013-12-18
dc.identifier.citationAnthony , T , Polani , D & Nehaniv , C L 2013 , ' General self-motivation and strategy identification : Case studies based on Sokoban and Pac-Man ' , IEEE Transactions on Computational Intelligence and AI in Games , vol. 6 , no. 1 , 6687219 , pp. 1-17 . https://doi.org/10.1109/TCIAIG.2013.2295372
dc.identifier.issn1943-068X
dc.identifier.otherPURE: 8069328
dc.identifier.otherPURE UUID: 89375892-93b1-4414-943c-ed4f3dd6aba7
dc.identifier.otherScopus: 84896928584
dc.identifier.urihttp://hdl.handle.net/2299/15376
dc.description(c) 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
dc.description.abstractIn this paper, we use empowerment, a recently introduced biologically inspired measure, to allow an AI player to assign utility values to potential future states within a previously unencountered game without requiring explicit specification of goal states. We further introduce strategic affinity, a method of grouping action sequences together to form "strategies," by examining the overlap in the sets of potential future states following each such action sequence. We also demonstrate an information-theoretic method of predicting future utility. Combining these methods, we extend empowerment to soft-horizon empowerment which enables the player to select a repertoire of action sequences that aim to maintain anticipated utility. We show how this method provides a proto-heuristic for nonterminal states prior to specifying concrete game goals, and propose it as a principled candidate model for "intuitive" strategy selection, in line with other recent work on "self-motivated agent behavior." We demonstrate that the technique, despite being generically defined independently of scenario, performs quite well in relatively disparate scenarios, such as a Sokoban-inspired box-pushing scenario and in a Pac-Man-inspired predator game, suggesting novel and principle-based candidate routes toward more general game-playing algorithms.en
dc.format.extent17
dc.language.isoeng
dc.relation.ispartofIEEE Transactions on Computational Intelligence and AI in Games
dc.subjectArtificial intelligence (AI)
dc.subjectgames
dc.subjectinformation theory
dc.subjectArtificial Intelligence
dc.subjectSoftware
dc.subjectControl and Systems Engineering
dc.subjectElectrical and Electronic Engineering
dc.titleGeneral self-motivation and strategy identification : Case studies based on Sokoban and Pac-Manen
dc.contributor.institutionSchool of Computer Science
dc.contributor.institutionScience & Technology Research Institute
dc.contributor.institutionCentre for Computer Science and Informatics Research
dc.contributor.institutionAdaptive Systems
dc.description.statusPeer reviewed
rioxxterms.versionAM
rioxxterms.versionofrecordhttps://doi.org/10.1109/TCIAIG.2013.2295372
rioxxterms.typeJournal Article/Review
herts.preservation.rarelyaccessedtrue


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record