Show simple item record

dc.contributor.authorCsaji, BC
dc.contributor.authorMonostori, L
dc.date.available2014-05-21T22:51:39Z
dc.date.issued2008-01-01
dc.identifierhttp://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000257103000003&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=d4d813f4571fa7d6246bdc0dfeca3a1c
dc.identifier.citationCsaji, B. C. & Monostori, L. (2008). Adaptive stochastic resource control: A machine learning approach. JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH, 32, pp.453-486. https://doi.org/10.1613/jair.2548.
dc.identifier.issn1076-9757
dc.identifier.urihttp://hdl.handle.net/11343/29300
dc.description.abstractThe paper investigates stochastic resource allocation problems with scarce, reusable resources and non-preemtive, time-dependent, interconnected tasks. This approach is a natural generalization of several standard resource management problems, such as scheduling and transportation problems. First, reactive solutions are considered and defined as control policies of suitably reformulated Markov decision processes (MDPs). We argue that this reformulation has several favorable properties, such as it has finite state and action spaces, it is aperiodic, hence all policies are proper and the space of control policies can be safely restricted. Next, approximate dynamic programming (ADP) methods, such as fitted Q-learning, are suggested for computing an efficient control policy. In order to compactly maintain the cost-to-go function, two representations are studied: hash tables and support vector regression (SVR), particularly, nu-SVRs. Several additional improvements, such as the application of limited-lookahead rollout algorithms in the initial phases, action space decomposition, task clustering and distributed sampling are investigated, too. Finally, experimental results on both benchmark and industry-related data are presented.
dc.languageEnglish
dc.publisherAI ACCESS FOUNDATION
dc.subjectArtificial Intelligence and Image Processing
dc.titleAdaptive stochastic resource control: A machine learning approach
dc.typeJournal Article
dc.identifier.doi10.1613/jair.2548
melbourne.peerreviewPeer Reviewed
melbourne.affiliationThe University of Melbourne
melbourne.affiliation.departmentElectrical and Electronic Engineering
melbourne.source.titleJournal of Artificial Intelligence Research
melbourne.source.volume32
melbourne.source.pages453-486
dc.description.pagestart453
melbourne.publicationid166877
melbourne.elementsid337062
melbourne.contributor.authorCSAJI, BALAZS
dc.identifier.eissn1943-5037
melbourne.accessrightsThis item is currently not available from this repository


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