Publication:
A Scoring Method for Interpretability of Concepts in Convolutional Neural Networks, Evrişimsel Sinir Aǧlarinda Kavram Yorumlama için bir Puanlama Yöntemi

Simple item page
dc.contributor.authorGurkan, Mustafa Kaǧan
dc.contributor.authorArica, Nafiz
dc.contributor.authorVural, Fatos T.Yarman
dc.contributor.institutionGurkan, Mustafa Kaǧan, Bahçeşehir Üniversitesi, Istanbul, Turkey
dc.contributor.institutionArica, Nafiz, Bahçeşehir Üniversitesi, Istanbul, Turkey
dc.contributor.institutionVural, Fatos T.Yarman, Middle East Technical University (METU), Ankara, Turkey
dc.date.accessioned2025-10-05T15:22:52Z
dc.date.issued2022
dc.description.abstractIn this paper, we propose a scoring algorithm for measuring the interpretability of CNN models by focusing on the feature extraction operation at the convolutional layers. The proposed approach is based on the principal of concept analysis, for a predefined list of concepts. A map of the network is created based on its responsiveness against each concept. Once this map is ready, various images can be applied as inputs and they are matched with the concepts whose hidden nodes are highly activated. Finally, the evaluation algorithm kicks in to use these descriptions during the final prediction and provides human-understandable explanations. © 2022 Elsevier B.V., All rights reserved.
dc.identifier.conferenceName30th Signal Processing and Communications Applications Conference, SIU 2022
dc.identifier.conferencePlaceSafranbolu
dc.identifier.doi10.1109/SIU55565.2022.9864930
dc.identifier.isbn9781665450928
dc.identifier.scopus2-s2.0-85138693532
dc.identifier.urihttps://doi.org/10.1109/SIU55565.2022.9864930
dc.identifier.urihttps://hdl.handle.net/20.500.14719/9074
dc.language.isotr
dc.publisherInstitute of Electrical and Electronics Engineers Inc.
dc.subject.authorkeywordsConcept-based Analysis
dc.subject.authorkeywordsConvolutional Neural Networks
dc.subject.authorkeywordsExplainable Ai
dc.subject.authorkeywordsRuntime Interpretation
dc.subject.authorkeywordsConvolution
dc.subject.authorkeywordsCnn Models
dc.subject.authorkeywordsConcept-based
dc.subject.authorkeywordsConcept-based Analyse
dc.subject.authorkeywordsConvolutional Neural Network
dc.subject.authorkeywordsExplainable Ai
dc.subject.authorkeywordsInterpretability
dc.subject.authorkeywordsRuntime Interpretation
dc.subject.authorkeywordsRuntimes
dc.subject.authorkeywordsScoring Algorithms
dc.subject.authorkeywordsScoring Methods
dc.subject.authorkeywordsConvolutional Neural Networks
dc.subject.indexkeywordsConvolution
dc.subject.indexkeywordsCNN models
dc.subject.indexkeywordsConcept-based
dc.subject.indexkeywordsConcept-based analyse
dc.subject.indexkeywordsConvolutional neural network
dc.subject.indexkeywordsExplainable AI
dc.subject.indexkeywordsInterpretability
dc.subject.indexkeywordsRuntime interpretation
dc.subject.indexkeywordsRuntimes
dc.subject.indexkeywordsScoring algorithms
dc.subject.indexkeywordsScoring methods
dc.subject.indexkeywordsConvolutional neural networks
dc.titleA Scoring Method for Interpretability of Concepts in Convolutional Neural Networks, Evrişimsel Sinir Aǧlarinda Kavram Yorumlama için bir Puanlama Yöntemi
dc.typeConference Paper
dcterms.referencesGrad Cam Visual Explanations from Deep Networks Via Gradient Based Localization, (2016), Interpretability Beyond Feature Attribution Quantitative Testing with Concept Activation Vectors Tcav, (2017), Towards A Rigorous Science of Interpretable Machine Learning, (2017), Bach, Sebastian, On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation, PLOS ONE, 10, 7, (2015), Zhou, Bolei, Learning Deep Features for Discriminative Localization, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016-December, pp. 2921-2929, (2016), Deep Image Prior, (2017), Bau, David Anthony, Network dissection: Quantifying interpretability of deep visual representations, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2017-January, pp. 3319-3327, (2017), Bau, David Anthony, Understanding the role of individual units in a deep neural network, Proceedings of the National Academy of Sciences of the United States of America, 117, 48, pp. 30071-30078, (2020)
dspace.entity.typePublication
local.indexed.atScopus
person.identifier.scopus-author-id57479133200
person.identifier.scopus-author-id56247026400
person.identifier.scopus-author-id57209180291

Files

Collections

Scopus İndeksli Yayınlar Koleksiyonu