Training Optimum-Path Forest on Graphics Processing Units

Página do item simplificado
dc.contributor.authorIwashita, Adriana S. [UNESP]
dc.contributor.authorRomero, Marcos V. T. [UNESP]
dc.contributor.authorBaldassin, Alexandro [UNESP]
dc.contributor.authorCosta, Kelton A. P. [UNESP]
dc.contributor.authorPapa, Joao P. [UNESP]
dc.contributor.authorBattiato, S.
dc.contributor.authorBraz, J.
dc.contributor.institutionUniversidade Estadual Paulista (Unesp)
dc.date.accessioned2019-10-04T12:30:15Z
dc.date.available2019-10-04T12:30:15Z
dc.date.issued2014-01-01
dc.description.abstractIn this paper, we presented a Graphics Processing Unit (GPU)-based training algorithm for Optimum-Path Forest (OPF) classifier. The proposed approach employs the idea of a vector-matrix multiplication to speed up both traditional OPF training algorithm and a recently proposed Central Processing Unit (CPU)-based OPF training algorithm. Experiments in several public datasets have showed the efficiency of the proposed approach, which demonstrated to be up to 14 times faster for some datasets. To the best of our knowledge, this is the first GPU-based implementation for OPF training algorithm.en
dc.description.affiliationSao Paulo State Univ, Dept Comp, Bauru, SP, Brazil
dc.description.affiliationSao Paulo State Univ, Dept Stat Appl Math & Computat, Rio Claro, SP, Brazil
dc.description.affiliationUnespSao Paulo State Univ, Dept Comp, Bauru, SP, Brazil
dc.description.affiliationUnespSao Paulo State Univ, Dept Stat Appl Math & Computat, Rio Claro, SP, Brazil
dc.description.sponsorshipFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
dc.description.sponsorshipConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.description.sponsorshipIdFAPESP: 2009/16206-1
dc.description.sponsorshipIdFAPESP: 2010/12697-8
dc.description.sponsorshipIdFAPESP: 2011/08348-0
dc.description.sponsorshipIdCNPq: 470571/2013-6
dc.description.sponsorshipIdCNPq: 303182/2011-3
dc.format.extent581-588
dc.identifier.citationProceedings Of The 2014 9th International Conference On Computer Vision, Theory And Applications (visapp 2014), Vol 2. New York: Ieee, p. 581-588, 2014.
dc.identifier.urihttp://hdl.handle.net/11449/184816
dc.identifier.wosWOS:000412737200071
dc.language.isoeng
dc.publisherIeee
dc.relation.ispartofProceedings Of The 2014 9th International Conference On Computer Vision, Theory And Applications (visapp 2014), Vol 2
dc.rights.accessRightsAcesso aberto
dc.sourceWeb of Science
dc.subjectOptimum-Path Forest
dc.subjectGraphics Processing Unit
dc.titleTraining Optimum-Path Forest on Graphics Processing Unitsen
dc.typeTrabalho apresentado em evento
dcterms.licensehttp://www.ieee.org/publications_standards/publications/rights/rights_policies.html
dcterms.rightsHolderIeee
unesp.author.lattes4738829911864396[3]
unesp.author.orcid0000-0001-8824-3055[3]
unesp.campusUniversidade Estadual Paulista (Unesp), Faculdade de Ciências, Baurupt
unesp.departmentComputação - FCpt

Arquivos

Coleções

Bauru - FC - Faculdade de Ciências
Artigos - Estatistica, Matemática Aplicada e Computação - IGCE