Afficher la notice abrégée

Evolving Genetic Programming Classifiers with Novelty Search

hal.structure.identifierInstituto Tecnológico de Tijuana = Tijuana Institute of Technology [Tijuana]
dc.contributor.authorNAREDO, Enrique
hal.structure.identifierInstituto Tecnológico de Tijuana = Tijuana Institute of Technology [Tijuana]
dc.contributor.authorTRUJILLO, Leonardo
hal.structure.identifierInstitut de Mathématiques de Bordeaux [IMB]
hal.structure.identifierQuality control and dynamic reliability [CQFD]
dc.contributor.authorLEGRAND, Pierrick
hal.structure.identifierInstituto de Engenharia de Sistemas e Computadores [INESC]
dc.contributor.authorSILVA, Sara
hal.structure.identifierInstituto Tecnológico de Tijuana = Tijuana Institute of Technology [Tijuana]
dc.contributor.authorMUNOZ, Luis
dc.date.accessioned2024-04-04T03:13:20Z
dc.date.available2024-04-04T03:13:20Z
dc.date.issued2016-11-10
dc.identifier.issn0020-0255
dc.identifier.urihttps://oskar-bordeaux.fr/handle/20.500.12278/193943
dc.description.abstractEnNovelty Search (NS) is a unique approach towards search and optimization, where an explicit objective function is replaced by a measure of solution novelty. However, NS has been mostly used in evolutionary robotics while its usefulness in classic machine learning problems has not been explored. This work presents a NS-based genetic programming (GP) algorithm for supervised classification. Results show that NS can solve real-world classification tasks, the algorithm is validated on real-world benchmarks for binary and multiclass problems. These results are made possible by using a domain-specific behavior descriptor. Moreover, two new versions of the NS algorithm are proposed, Probabilistic NS (PNS) and a variant of Minimal Criteria NS (MCNS). The former models the behavior of each solution as a random vector and eliminates all of the original NS parameters while reducing the computational overhead of the NS algorithm. The latter uses a standard objective function to constrain and bias the search towards high performance solutions. The paper also discusses the effects of NS on GP search dynamics and code growth. Results show that NS can be used as a realistic alternative for supervised classification, and specifically for binary problems the NS algorithm exhibits an implicit bloat control ability.
dc.language.isoen
dc.publisherElsevier
dc.title.enEvolving Genetic Programming Classifiers with Novelty Search
dc.typeArticle de revue
dc.identifier.doi10.1016/j.ins.2016.06.044
dc.subject.halInformatique [cs]/Intelligence artificielle [cs.AI]
dc.description.sponsorshipEuropeAnalysis and classification of mental states of vigilance with evolutionary computation
bordeaux.journalInformation Sciences
bordeaux.page347–367
bordeaux.volume369
bordeaux.hal.laboratoriesInstitut de Mathématiques de Bordeaux (IMB) - UMR 5251*
bordeaux.institutionUniversité de Bordeaux
bordeaux.institutionBordeaux INP
bordeaux.institutionCNRS
bordeaux.peerReviewedoui
hal.identifierhal-01389049
hal.version1
hal.popularnon
hal.audienceInternationale
hal.origin.linkhttps://hal.archives-ouvertes.fr//hal-01389049v1
bordeaux.COinSctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.jtitle=Information%20Sciences&rft.date=2016-11-10&rft.volume=369&rft.spage=347%E2%80%93367&rft.epage=347%E2%80%93367&rft.eissn=0020-0255&rft.issn=0020-0255&rft.au=NAREDO,%20Enrique&TRUJILLO,%20Leonardo&LEGRAND,%20Pierrick&SILVA,%20Sara&MUNOZ,%20Luis&rft.genre=article


Fichier(s) constituant ce document

FichiersTailleFormatVue

Il n'y a pas de fichiers associés à ce document.

Ce document figure dans la(les) collection(s) suivante(s)

Afficher la notice abrégée