Publicação

Genetic programming with semantic equivalence classes

Ver documento

Detalhes bibliográficos
Resumo:In this paper, we introduce the concept of semantics-based equivalence classes for symbolic regression problems in genetic programming. The idea is implemented by means of two different genetic programming systems, in which two different definitions of equivalence are used. In both systems, whenever a solution in an equivalence class is found, it is possible to generate any other solution in that equivalence class analytically. As such, these two systems allow us to shift the objective of genetic programming: instead of finding a globally optimal solution, the objective is now to find any solution that belongs to the same equivalence class as a global optimum. Further, we propose improvements to these genetic programming systems in which, once a solution that belongs to a particular equivalence class is generated, no other solution in that class is accepted in the population during the evolution anymore. We call these improved versions filtered systems. Experimental results obtained via seven complex real-life test problems show that using equivalence classes is a promising idea and that filters are generally helpful for improving the systems' performance. Furthermore, the proposed methods produce individuals with a much smaller size with respect to geometric semantic genetic programming. Finally, we show that filters are also useful to improve the performance of a state-of-the-art method, not explicitly based on semantic equivalence classes, like linear scaling.
Autores principais:Ruberto, Stefano
Outros Autores:Vanneschi, Leonardo; Castelli, Mauro
Assunto:Equivalence classes Genetic programming Semantics General Computer Science General Mathematics
Ano:2019
País:Portugal
Tipo de documento:artigo
Tipo de acesso:acesso aberto
Instituição associada:Universidade Nova de Lisboa
Idioma:inglês
Origem:Repositório Institucional da UNL
_version_ 1865920626398593024
author Ruberto, Stefano
author2 Vanneschi, Leonardo
Castelli, Mauro
author2_role author
author
author_facet Ruberto, Stefano
Ruberto, Stefano
Vanneschi, Leonardo
Castelli, Mauro
Vanneschi, Leonardo
Castelli, Mauro
author_role author
contributor_name_str_mv NOVA Information Management School (NOVA IMS)
Information Management Research Center (MagIC) - NOVA Information Management School
Elsevier
RUN
country_str PT
creators_json_str [{\"Person.name\":\"Ruberto, Stefano\"},{\"Person.name\":\"Vanneschi, Leonardo\"},{\"Person.name\":\"Castelli, Mauro\"}]
datacite.contributors.contributor.contributorName.fl_str_mv NOVA Information Management School (NOVA IMS)
Information Management Research Center (MagIC) - NOVA Information Management School
Elsevier
RUN
datacite.creators.creator.creatorName.fl_str_mv Ruberto, Stefano
Vanneschi, Leonardo
Castelli, Mauro
datacite.date.Accepted.fl_str_mv 2019-02-01T00:00:00Z
datacite.date.available.fl_str_mv 2024-01-27T01:32:02Z
datacite.date.embargoed.fl_str_mv 2024-01-27T01:32:02Z
datacite.rights.fl_str_mv http://purl.org/coar/access_right/c_abf2
datacite.subjects.subject.fl_str_mv Equivalence classes
Genetic programming
Semantics
General Computer Science
General Mathematics
datacite.titles.title.fl_str_mv Genetic programming with semantic equivalence classes
dc.contributor.none.fl_str_mv NOVA Information Management School (NOVA IMS)
Information Management Research Center (MagIC) - NOVA Information Management School
Elsevier
RUN
dc.creator.none.fl_str_mv Ruberto, Stefano
Vanneschi, Leonardo
Castelli, Mauro
dc.date.Accepted.fl_str_mv 2019-02-01T00:00:00Z
dc.date.available.fl_str_mv 2024-01-27T01:32:02Z
dc.date.embargoed.fl_str_mv 2024-01-27T01:32:02Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv http://hdl.handle.net/10362/151422
dc.language.none.fl_str_mv eng
dc.rights.none.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.subject.none.fl_str_mv Equivalence classes
Genetic programming
Semantics
General Computer Science
General Mathematics
dc.title.fl_str_mv Genetic programming with semantic equivalence classes
dc.type.none.fl_str_mv http://purl.org/coar/resource_type/c_6501
description In this paper, we introduce the concept of semantics-based equivalence classes for symbolic regression problems in genetic programming. The idea is implemented by means of two different genetic programming systems, in which two different definitions of equivalence are used. In both systems, whenever a solution in an equivalence class is found, it is possible to generate any other solution in that equivalence class analytically. As such, these two systems allow us to shift the objective of genetic programming: instead of finding a globally optimal solution, the objective is now to find any solution that belongs to the same equivalence class as a global optimum. Further, we propose improvements to these genetic programming systems in which, once a solution that belongs to a particular equivalence class is generated, no other solution in that class is accepted in the population during the evolution anymore. We call these improved versions filtered systems. Experimental results obtained via seven complex real-life test problems show that using equivalence classes is a promising idea and that filters are generally helpful for improving the systems' performance. Furthermore, the proposed methods produce individuals with a much smaller size with respect to geometric semantic genetic programming. Finally, we show that filters are also useful to improve the performance of a state-of-the-art method, not explicitly based on semantic equivalence classes, like linear scaling.
dirty 0
eu_rights_str_mv openAccess
format article
fulltext.url.fl_str_mv https://run.unl.pt/bitstreams/21747b94-066f-4441-beeb-891f6e26c3f5/download
id run_e89a4ed9a8df3bbe736f80d4eaa5613d
identifier.url.fl_str_mv http://hdl.handle.net/10362/151422
instacron_str unl
institution Universidade Nova de Lisboa
instname_str Universidade Nova de Lisboa
language eng
network_acronym_str run
network_name_str Repositório Institucional da UNL
oai_identifier_str oai:run.unl.pt:10362/151422
organization_str_mv urn:organizationAcronym:unl
person_str_mv Ruberto, Stefano
Vanneschi, Leonardo
Castelli, Mauro
publishDate 2019
reponame_str Repositório Institucional da UNL
repository_id_str urn:repositoryAcronym:run
service_str_mv urn:repositoryAcronym:run
spelling engenIn this paper, we introduce the concept of semantics-based equivalence classes for symbolic regression problems in genetic programming. The idea is implemented by means of two different genetic programming systems, in which two different definitions of equivalence are used. In both systems, whenever a solution in an equivalence class is found, it is possible to generate any other solution in that equivalence class analytically. As such, these two systems allow us to shift the objective of genetic programming: instead of finding a globally optimal solution, the objective is now to find any solution that belongs to the same equivalence class as a global optimum. Further, we propose improvements to these genetic programming systems in which, once a solution that belongs to a particular equivalence class is generated, no other solution in that class is accepted in the population during the evolution anymore. We call these improved versions filtered systems. Experimental results obtained via seven complex real-life test problems show that using equivalence classes is a promising idea and that filters are generally helpful for improving the systems' performance. Furthermore, the proposed methods produce individuals with a much smaller size with respect to geometric semantic genetic programming. Finally, we show that filters are also useful to improve the performance of a state-of-the-art method, not explicitly based on semantic equivalence classes, like linear scaling.application/pdfenGenetic programming with semantic equivalence classesRuberto, StefanoVanneschi, LeonardoCastelli, MauroNOVA Information Management School (NOVA IMS)Information Management Research Center (MagIC) - NOVA Information Management SchoolElsevierHostingInstitutionOrganizationalRUNe-mailmailto:run@unl.ptrun@unl.ptISSNIsPartOf2210-6502URNIsPartOfPURE: 5097819URNIsPartOfPURE UUID: 6c83e125-f582-4e11-af92-99393889f5bfURNIsPartOfScopus: 85048885945URNIsPartOfWOS: 000456761600032URNIsPartOfORCID: /0000-0002-8793-1451/work/131992546URNIsPartOfORCID: /0000-0003-4732-3328/work/151426701DOIIsPartOf10.1016/j.swevo.2018.06.0012024-01-27T01:32:02Z2019-022019-02-01T00:00:00ZHandlehttp://hdl.handle.net/10362/151422http://purl.org/coar/access_right/c_abf2open accessEquivalence classesGenetic programmingSemanticsGeneral Computer ScienceGeneral Mathematics830260 bytesliteraturehttp://purl.org/coar/resource_type/c_6501journal articlehttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://run.unl.pt/bitstreams/21747b94-066f-4441-beeb-891f6e26c3f5/download
spellingShingle Genetic programming with semantic equivalence classes
Genetic programming with semantic equivalence classes
Ruberto, Stefano
Equivalence classes
Genetic programming
Semantics
General Computer Science
General Mathematics
Ruberto, Stefano
Equivalence classes
Genetic programming
Semantics
General Computer Science
General Mathematics
status NEW
subject.fl_str_mv Equivalence classes
Genetic programming
Semantics
General Computer Science
General Mathematics
title Genetic programming with semantic equivalence classes
title_full Genetic programming with semantic equivalence classes
title_fullStr Genetic programming with semantic equivalence classes
Genetic programming with semantic equivalence classes
title_full_unstemmed Genetic programming with semantic equivalence classes
Genetic programming with semantic equivalence classes
title_short Genetic programming with semantic equivalence classes
title_sort Genetic programming with semantic equivalence classes
topic Equivalence classes
Genetic programming
Semantics
General Computer Science
General Mathematics
topic_facet Equivalence classes
Genetic programming
Semantics
General Computer Science
General Mathematics
url http://hdl.handle.net/10362/151422
visible 1