Show simple item record

dc.contributor.advisorMiikkulainen, Ristoen
dc.creatorLockett, Alan Justinen
dc.date.accessioned2012-07-05T19:50:55Zen
dc.date.available2012-07-05T19:50:55Zen
dc.date.issued2012-05en
dc.date.submittedMay 2012en
dc.identifier.urihttp://hdl.handle.net/2152/ETD-UT-2012-05-5459en
dc.descriptiontexten
dc.description.abstractThe primary goal of artificial intelligence research is to develop a machine capable of learning to solve disparate real-world tasks autonomously, without relying on specialized problem-specific inputs. This dissertation suggests that such machines are realistic: If No Free Lunch theorems were to apply to all real-world problems, then the world would be utterly unpredictable. In response, the dissertation proposes the information-maximization principle, which claims that the optimal optimization methods make the best use of the information available to them. This principle results in a new algorithm, evolutionary annealing, which is shown to perform well especially in challenging problems with irregular structure.en
dc.format.mimetypeapplication/pdfen
dc.language.isoengen
dc.subjectOptimizationen
dc.subjectGeneral-purpose learningen
dc.subjectMartingale optimizationen
dc.subjectArtificial intelligenceen
dc.subjectEvolutionary computationen
dc.subjectGenetic algorithmsen
dc.subjectSimulated annealingen
dc.subjectEvolutionary annealingen
dc.subjectNeuroannealingen
dc.subjectNeural networksen
dc.subjectNeural network controllersen
dc.subjectNeuroevolutionen
dc.subjectDifferential evolutionen
dc.subjectNo Free Lunch theoremsen
dc.subjectNFL Identification Theoremen
dc.subjectPopulation-based stochastic optimizationen
dc.subjectIterative optimizationen
dc.subjectOptimal optimizationen
dc.subjectInformation-maximization principleen
dc.subjectConvex controlen
dc.subjectAlgorithm selectionen
dc.titleGeneral-purpose optimization through information maximizationen
dc.date.updated2012-07-05T19:51:10Zen
dc.identifier.slug2152/ETD-UT-2012-05-5459en
dc.contributor.committeeMemberGhosh, Joydeepen
dc.contributor.committeeMemberMooney, Raymonden
dc.contributor.committeeMemberRavikumar, Pradeepen
dc.contributor.committeeMemberZitkovic, Gordanen
dc.description.departmentComputer Sciencesen
dc.type.genrethesisen
thesis.degree.departmentComputer Sciencesen
thesis.degree.disciplineComputer Scienceen
thesis.degree.grantorUniversity of Texas at Austinen
thesis.degree.levelDoctoralen
thesis.degree.nameDoctor of Philosophyen


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record