Discriminative sparse coding on multi-manifolds

Handle URI:
http://hdl.handle.net/10754/334545
Title:
Discriminative sparse coding on multi-manifolds
Authors:
Wang, J.J.-Y.; Bensmail, H.; Yao, N.; Gao, Xin ( 0000-0002-7108-3574 )
Abstract:
Sparse coding has been popularly used as an effective data representation method in various applications, such as computer vision, medical imaging and bioinformatics. However, the conventional sparse coding algorithms and their manifold-regularized variants (graph sparse coding and Laplacian sparse coding), learn codebooks and codes in an unsupervised manner and neglect class information that is available in the training set. To address this problem, we propose a novel discriminative sparse coding method based on multi-manifolds, that learns discriminative class-conditioned codebooks and sparse codes from both data feature spaces and class labels. First, the entire training set is partitioned into multiple manifolds according to the class labels. Then, we formulate the sparse coding as a manifold-manifold matching problem and learn class-conditioned codebooks and codes to maximize the manifold margins of different classes. Lastly, we present a data sample-manifold matching-based strategy to classify the unlabeled data samples. Experimental results on somatic mutations identification and breast tumor classification based on ultrasonic images demonstrate the efficacy of the proposed data representation and classification approach. 2013 The Authors. All rights reserved.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computational Bioscience Research Center (CBRC)
Citation:
Wang JJ-Y, Bensmail H, Yao N, Gao X (2013) Discriminative sparse coding on multi-manifolds. Knowledge-Based Systems 54: 199-206. doi:10.1016/j.knosys.2013.09.004.
Publisher:
Elsevier BV
Journal:
Knowledge-Based Systems
Issue Date:
26-Sep-2013
DOI:
10.1016/j.knosys.2013.09.004
Type:
Article
ISSN:
09507051
Appears in Collections:
Articles; Computational Bioscience Research Center (CBRC); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorWang, J.J.-Y.en
dc.contributor.authorBensmail, H.en
dc.contributor.authorYao, N.en
dc.contributor.authorGao, Xinen
dc.date.accessioned2014-11-11T14:29:09Z-
dc.date.available2014-11-11T14:29:09Z-
dc.date.issued2013-09-26en
dc.identifier.citationWang JJ-Y, Bensmail H, Yao N, Gao X (2013) Discriminative sparse coding on multi-manifolds. Knowledge-Based Systems 54: 199-206. doi:10.1016/j.knosys.2013.09.004.en
dc.identifier.issn09507051en
dc.identifier.doi10.1016/j.knosys.2013.09.004en
dc.identifier.urihttp://hdl.handle.net/10754/334545en
dc.description.abstractSparse coding has been popularly used as an effective data representation method in various applications, such as computer vision, medical imaging and bioinformatics. However, the conventional sparse coding algorithms and their manifold-regularized variants (graph sparse coding and Laplacian sparse coding), learn codebooks and codes in an unsupervised manner and neglect class information that is available in the training set. To address this problem, we propose a novel discriminative sparse coding method based on multi-manifolds, that learns discriminative class-conditioned codebooks and sparse codes from both data feature spaces and class labels. First, the entire training set is partitioned into multiple manifolds according to the class labels. Then, we formulate the sparse coding as a manifold-manifold matching problem and learn class-conditioned codebooks and codes to maximize the manifold margins of different classes. Lastly, we present a data sample-manifold matching-based strategy to classify the unlabeled data samples. Experimental results on somatic mutations identification and breast tumor classification based on ultrasonic images demonstrate the efficacy of the proposed data representation and classification approach. 2013 The Authors. All rights reserved.en
dc.language.isoenen
dc.publisherElsevier BVen
dc.rightshttp://creativecommons.org/licenses/by-nc-sa/3.0/en
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/3.0/en
dc.subjectData representationen
dc.subjectLarge marginsen
dc.subjectMulti-manifoldsen
dc.subjectSparse codingen
dc.subjectBioinformaticsen
dc.subjectMedical imagingen
dc.subjectUltrasonic imagingen
dc.subjectBreast tumor classificationsen
dc.subjectClass informationen
dc.subjectClassification approachen
dc.subjectData representationsen
dc.subjectMatching problemsen
dc.subjectCodes (symbols)en
dc.titleDiscriminative sparse coding on multi-manifoldsen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentComputational Bioscience Research Center (CBRC)en
dc.identifier.journalKnowledge-Based Systemsen
dc.eprint.versionPublisher's Version/PDFen
dc.contributor.institutionQatar Computing Research Institute, Doha 5825, Qataren
dc.contributor.institutionDepartment of Instrumentation Engineering, Shanghai Jiao Tong University, Shanghai 200240, Chinaen
dc.contributor.institutionNational Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210023, Chinaen
dc.contributor.affiliationKing Abdullah University of Science and Technology (KAUST)en
kaust.authorWang, Jim Jing-Yanen
kaust.authorGao, Xinen
This item is licensed under a Creative Commons License
Creative Commons
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.