Vis enkel innførsel

dc.contributor.authorJenssen, Robert
dc.contributor.authorKloft, Marius
dc.contributor.authorZien, Alexander
dc.contributor.authorSonnenburg, Soeren
dc.contributor.authorMüller, Klaus R.
dc.date.accessioned2013-03-13T07:30:48Z
dc.date.available2013-03-13T07:30:48Z
dc.date.issued2012
dc.description.abstractWe provide a novel interpretation of the dual of support vector machines (SVMs) in terms of scatter with respect to class prototypes and their mean. As a key contribution, we extend this framework to multiple classes, providing a new joint Scatter SVM algorithm, at the level of its binary counterpart in the number of optimization variables. This enables us to implement computationally efficient solvers based on sequential minimal and chunking optimization. As a further contribution, the primal problem formulation is developed in terms of regularized risk minimization and the hinge loss, revealing the score function to be used in the actual classification of test patterns. We investigate Scatter SVM properties related to generalization ability, computational efficiency, sparsity and sensitivity maps, and report promising results.en
dc.identifier.citationPLoS ONE (2012), vol.7(10): e42947en
dc.identifier.cristinIDFRIDAID 989991
dc.identifier.doihttp://dx.doi.org/10.1371/journal.pone.0042947
dc.identifier.issn1932-6203
dc.identifier.urihttps://hdl.handle.net/10037/4963
dc.identifier.urnURN:NBN:no-uit_munin_4698
dc.language.isoengen
dc.publisherPublic Library of Science (PLoS)en
dc.rights.accessRightsopenAccess
dc.subjectVDP::Mathematics and natural science: 400::Information and communication science: 420::Knowledge based systems: 425en
dc.subjectVDP::Matematikk og Naturvitenskap: 400::Informasjons- og kommunikasjonsvitenskap: 420::Kunnskapsbaserte systemer: 425en
dc.subjectVDP::Mathematics and natural science: 400::Information and communication science: 420::Simulation, visualization, signal processing, image processing: 429en
dc.subjectVDP::Matematikk og Naturvitenskap: 400::Informasjons- og kommunikasjonsvitenskap: 420::Simulering, visualisering, signalbehandling, bildeanalyse: 429en
dc.titleA Scatter-Based Prototype Framework and Multi-Class Extension of Support Vector Machinesen
dc.typeJournal articleen
dc.typeTidsskriftartikkelen
dc.typePeer revieweden


Tilhørende fil(er)

Thumbnail
Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel