Show simple item record

dc.contributor.authorWang, Jim Jing-Yan
dc.contributor.authorWang, Yunji
dc.contributor.authorJing, Bing-Yi
dc.contributor.authorGao, Xin
dc.date.accessioned2015-09-15T07:15:42Z
dc.date.available2015-09-15T07:15:42Z
dc.date.issued2015-02-12
dc.identifier.citationRegularized maximum correntropy machine 2015, 160:85 Neurocomputing
dc.identifier.issn09252312
dc.identifier.doi10.1016/j.neucom.2014.09.080
dc.identifier.urihttp://hdl.handle.net/10754/577311
dc.description.abstractIn this paper we investigate the usage of regularized correntropy framework for learning of classifiers from noisy labels. The class label predictors learned by minimizing transitional loss functions are sensitive to the noisy and outlying labels of training samples, because the transitional loss functions are equally applied to all the samples. To solve this problem, we propose to learn the class label predictors by maximizing the correntropy between the predicted labels and the true labels of the training samples, under the regularized Maximum Correntropy Criteria (MCC) framework. Moreover, we regularize the predictor parameter to control the complexity of the predictor. The learning problem is formulated by an objective function considering the parameter regularization and MCC simultaneously. By optimizing the objective function alternately, we develop a novel predictor learning algorithm. The experiments on two challenging pattern classification tasks show that it significantly outperforms the machines with transitional loss functions.
dc.language.isoen
dc.publisherElsevier BV
dc.relation.urlhttp://linkinghub.elsevier.com/retrieve/pii/S0925231215001150
dc.rightsNOTICE: this is the author’s version of a work that was accepted for publication in Neurocomputing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Neurocomputing, 12 February 2015. DOI: 10.1016/j.neucom.2014.09.080
dc.subjectPattern classification
dc.subjectLabel noise
dc.subjectMaximum Correntropy Criteria
dc.subjectRegularization
dc.titleRegularized maximum correntropy machine
dc.typeArticle
dc.contributor.departmentComputational Bioscience Research Center (CBRC)
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.identifier.journalNeurocomputing
dc.eprint.versionPost-print
dc.contributor.institutionElectrical and Computer Engineering Department, The University of Texas at San Antonio, San Antonio, TX 78249, USA
dc.contributor.institutionDepartment of Mathematics, Hong Kong University of Science and Technology, Kowloon, Hong Kong
dc.contributor.affiliationKing Abdullah University of Science and Technology (KAUST)
dc.identifier.arxivid1501.04282
kaust.personGao, Xin
kaust.personWang, Jim Jing-Yan
refterms.dateFOA2017-02-12T00:00:00Z
dc.date.published-online2015-02-12
dc.date.published-print2015-07


Files in this item

Thumbnail
Name:
RegMaxCEM_NEUCOM.pdf
Size:
169.3Kb
Format:
PDF
Description:
Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record