SSS-Net: A shadowed-sets-based semi-supervised sample selection network for classification on noise labeled images

dc.authorwosidzhang, hongyun/JBJ-7502-2023
dc.contributor.authorCai, Kecan
dc.contributor.authorZhang, Hongyun
dc.contributor.authorPedrycz, Witold
dc.contributor.authorMiao, Duoqian
dc.date.accessioned2024-05-19T14:46:38Z
dc.date.available2024-05-19T14:46:38Z
dc.date.issued2023
dc.departmentİstinye Üniversitesien_US
dc.description.abstractSample selection is a fundamental technique utilized in image classification with noisy labels. A plethora of sample selection approaches published in the literature are based on a small-loss strategy, in which division thresholds are set manually and the correlation between sample losses is ignored. Furthermore, one of the most evident shortcomings of these approaches is that noisy samples with low-quality pseudo-labels can negatively impact the model resulting in poor performance. In this study, a shadowed-sets-based semi-supervised sample selection network called SSS-Net is developed to address these limitations. Our approach leverages a novel technique that combines a loss-similarity-based-clustering method (LSCM) with the shadowed-sets theory to adaptively select clean samples. We then introduce an original high-quality pseudo-label sample reselection (HPSR) strategy, which is designed through the co-training of two networks, to pick the samples with high-quality pseudo-labels. Finally, the selected samples are utilized to further train the network and complete classification. This study presents an automated approach that determines optimal division thresholds to select clean samples adaptively. Furthermore, it improves the current semi-supervised sample selection method by effectively utilizing noisy samples. The suitability and promising performance of the proposed approach are supported through experimental studies using five real-world datasets. Comparative studies involving several state-of-the-art methods are also reported. & COPY; 2023 Elsevier B.V. All rights reserved.en_US
dc.description.sponsorshipNational Natural Science Foundation of China [62076182, 61976158, 61976160, 62076184]; Natural Science Foundation of Shanghai, China [22ZR1466700]en_US
dc.description.sponsorshipAcknowledgments We sincerely thank the editors and the anonymous review-ers for their valuable comments. This work was supported by the National Natural Science Foundation of China under Grants 62076182, 61976158, 61976160 and 62076184, and the Natural Science Foundation of Shanghai, China under Grant 22ZR1466700.en_US
dc.identifier.doi10.1016/j.knosys.2023.110732
dc.identifier.issn0950-7051
dc.identifier.issn1872-7409
dc.identifier.scopus2-s2.0-85164276193en_US
dc.identifier.scopusqualityQ1en_US
dc.identifier.urihttps://doi.org10.1016/j.knosys.2023.110732
dc.identifier.urihttps://hdl.handle.net/20.500.12713/5564
dc.identifier.volume276en_US
dc.identifier.wosWOS:001039258500001en_US
dc.identifier.wosqualityN/Aen_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherElsevieren_US
dc.relation.ispartofKnowledge-Based Systemsen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.snmz20240519_kaen_US
dc.subjectShadowed Setsen_US
dc.subjectAdaptive Division Thresholden_US
dc.subjectSemi-Supervised Learningen_US
dc.subjectImages Classificationen_US
dc.subjectNoisy Labelen_US
dc.titleSSS-Net: A shadowed-sets-based semi-supervised sample selection network for classification on noise labeled imagesen_US
dc.typeArticleen_US

Dosyalar