Frequency domain channel-wise attack to CNN classifiers in motor imagery brain-computer interfaces

Basit Öğe Kaydı
dc.authorscopusidWitold Pedrycz / 58861905800
dc.authorwosidWitold Pedrycz / FPE-7309-2022
dc.contributor.authorHuang, Xiuyu
dc.contributor.authorChoi, Kup-Sze
dc.contributor.authorLiang, Shuang
dc.contributor.authorZhang, Yuanpeng
dc.contributor.authorZhang, Yingkui
dc.contributor.authorPoon, Simon
dc.contributor.authorPedrycz, Witold
dc.date.accessioned2025-04-18T10:32:54Z
dc.date.available2025-04-18T10:32:54Z
dc.date.issued2024
dc.departmentİstinye Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Bilgisayar Mühendisliği Bölümü
dc.description.abstractObjective: Convolutional neural network (CNN), a classical structure in deep learning, has been commonly deployed in the motor imagery brain-computer interface (MIBCI). Many methods have been proposed to evaluate the vulnerability of such CNN models, primarily by attacking them using direct temporal perturbations. In this work, we propose a novel attacking approach based on perturbations in the frequency domain instead. Methods: For a given natural MI trial in the frequency domain, the proposed approach, called frequency domain channel-wise attack (FDCA), generates perturbations at each channel one after another to fool the CNN classifiers. The advances of this strategy are two-fold. First, instead of focusing on the temporal domain, perturbations are generated in the frequency domain where discriminative patterns can be extracted for motor imagery (MI) classification tasks. Second, the perturbing optimization is performed based on differential evolution algorithm in a black-box scenario where detailed model knowledge is not required. Results: Experimental results demonstrate the effectiveness of the proposed FDCA which achieves a significantly higher success rate than the baselines and existing methods in attacking three major CNN classifiers on four public MI benchmarks. Conclusion: Perturbations generated in the frequency domain yield highly competitive results in attacking MIBCI deployed by CNN models even in a black-box setting, where the model information is well-protected. Significance: To our best knowledge, existing MIBCI attack approaches are all gradient-based methods and require details about the victim model, e.g., the parameters and objective function. We provide a more flexible strategy that does not require model details but still produces an effective attack outcome.
dc.description.sponsorshipJiangsu University Philosophy and Social Science Foundation
dc.identifier.citationHuang, X., Choi, K. S., Liang, S., Zhang, Y., Zhang, Y., Poon, S., & Pedrycz, W. (2023). Frequency Domain Channel-wise Attack to CNN Classifiers in Motor Imagery Brain-Computer Interfaces. IEEE Transactions on Biomedical Engineering.
dc.identifier.doi10.1109/TBME.2023.3344295
dc.identifier.endpage1598
dc.identifier.issn0018-9294
dc.identifier.issn1558-2531
dc.identifier.issue5
dc.identifier.scopus2-s2.0-85181556090
dc.identifier.scopusqualityQ1
dc.identifier.startpage1587
dc.identifier.urihttp://dx.doi.org/10.1109/TBME.2023.3344295
dc.identifier.urihttps://hdl.handle.net/20.500.12713/7117
dc.identifier.volume71
dc.identifier.wosWOS:001262891800013
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.institutionauthorPedrycz, Pedrycz, Witold
dc.institutionauthoridWitold Pedrycz / 0000-0002-9335-9930
dc.language.isoen
dc.publisherIEEE-INST electronics electrical engineers
dc.relation.ispartofIEEE transactions on biomedical engineering
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.subjectAdversarial Attack
dc.subjectConvolutional Neural Networks
dc.subjectDifferential Evolution
dc.subjectFrequency Domain
dc.subjectMotor Imagery
dc.titleFrequency domain channel-wise attack to CNN classifiers in motor imagery brain-computer interfaces
dc.typeArticle

Dosyalar

Lisans paketi
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
license.txt
Boyut:
1.17 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

Bilgisayar Mühendisliği Bölümü Makale Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu