Efficient identification of improving moves in a ball for Pseudo-Boolean problems

dc.centroE.T.S.I. Informáticaes_ES
dc.contributor.authorChicano-García, José-Francisco
dc.contributor.authorWhitley, L. Darrell
dc.contributor.authorSutton, Andrew M.
dc.date.accessioned2014-06-27T08:34:32Z
dc.date.available2014-06-27T08:34:32Z
dc.date.created2014-06-26
dc.date.issued2014-06-27
dc.departamentoLenguajes y Ciencias de la Computación
dc.description.abstractHill climbing algorithms are at the core of many approaches to solve optimization problems. Such algorithms usually require the complete enumeration of a neighborhood of the current solution. In the case of problems defined over binary strings of length n, we define the r-ball neighborhood as the set of solutions at Hamming distance r or less from the current solution. For r << n this neighborhood contains Theta(nr) solutions. In this paper efficient methods are introduced to locate improving moves in the r-ball neighborhood for problems that can be written as a sum of a linear number of subfunctions depending on a bounded number of variables. NK-landscapes and MAX-kSAT are examples of these problems. If the number of subfunctions depending on any given variable is also bounded, then we prove that the method can explore the neighborhood in constant time, despite the fact that the number of solutions in the neighborhood is polynomial in n. We develop a hill climber based on our exploration method and we analyze its efficiency and efficacy using experiments with NKq-landscapes instances.es_ES
dc.description.sponsorshipUniversidad de Malaga. Campus de Excelencia Internacional Andalucia Tech. Ministerio de Educación (ayuda José Castillejo). Comisión Fulbright. Ministerio de Ciencia e Innovación y FEDER (TIN2011-28194). Contrato OTRI 8.06/5.47.4142 (VSB-Technical University of Ostrava). Air Force Office of Scientific Research, Air Force Materiel Command, USAF, (FA9550-11-1-0088).es_ES
dc.identifier.urihttp://hdl.handle.net/10630/7737
dc.language.isoenges_ES
dc.relation.eventdateJulio de 2014es_ES
dc.relation.eventplaceVancouver, Canadaes_ES
dc.relation.eventtitleGenetic and Evolutionary Computation Conferencees_ES
dc.rights.accessRightsopen accesses_ES
dc.subjectComputación evolutivaes_ES
dc.subject.otherNK-landscapeses_ES
dc.subject.otherPseudo-Boolean optimizationes_ES
dc.subject.otherHill climbinges_ES
dc.subject.otherLocal searches_ES
dc.titleEfficient identification of improving moves in a ball for Pseudo-Boolean problemses_ES
dc.typeconference outputes_ES
dspace.entity.typePublication
relation.isAuthorOfPublication6f65e289-6502-4756-871c-dbe0ca9be545
relation.isAuthorOfPublication.latestForDiscovery6f65e289-6502-4756-871c-dbe0ca9be545

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
pap429-chicano.pdf
Size:
332.04 KB
Format:
Adobe Portable Document Format