{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T07:10:50Z","timestamp":1777705850066,"version":"3.51.4"},"reference-count":19,"publisher":"SAGE Publications","issue":"6","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IFS"],"published-print":{"date-parts":[[2022,4,28]]},"abstract":"<jats:p>Universal Adversarial Perturbations(UAPs), which are image-agnostic adversarial perturbations, have been demonstrated to successfully deceive computer vision models. Proposed UAPs in the case of data-dependent, use the internal layers\u2019 activation or the output layer\u2019s decision values as supervision. In this paper, we use both of them to drive the supervised learning of UAP, termed as fully supervised UAP(FS-UAP), and design a progressive optimization strategy to solve the FS-UAP. Specifically, we define an internal layers supervised objective relying on multiple major internal layers\u2019 activation to estimate the deviations of adversarial examples from legitimate examples. We also define an output layer supervised objective relying on the logits of output layer to evaluate attacking degrees. In addition, we use the UAP found by previous stage as the initial solution of the next stage so as to progressively optimize the UAP stage-wise. We use seven networks and ImageNet dataset to evaluate the proposed FS-UAP, and provide an in-depth analysis for the latent factors affecting the performance of universal attacks. The experimental results show that our FS-UAP (i) has powerful capability of cheating CNNs (ii) has superior transfer-ability across models and weak data-dependent (iii) is appropriate for both untarget and target attacks.<\/jats:p>","DOI":"10.3233\/jifs-210728","type":"journal-article","created":{"date-parts":[[2021,12,24]],"date-time":"2021-12-24T10:26:30Z","timestamp":1640341590000},"page":"4959-4968","source":"Crossref","is-referenced-by-count":0,"title":["A fully supervised universal adversarial perturbations and the progressive optimization"],"prefix":"10.1177","volume":"42","author":[{"given":"Guangling","family":"Sun","sequence":"first","affiliation":[{"name":"School of Communication and Information Engineering, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haoqi","family":"Hu","sequence":"additional","affiliation":[{"name":"School of Communication and Information Engineering, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xinpeng","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Communication and Information Engineering, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaofeng","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Communication and Information Engineering, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","reference":[{"key":"10.3233\/JIFS-210728_ref1","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1016\/j.patrec.2018.10.003","article-title":"Joint spatial-spectral hyperspectral image classification based on convolutional neural network[J]","volume":"130","author":"Han","year":"2020","journal-title":"Pattern Recognition Letters"},{"key":"10.3233\/JIFS-210728_ref2","first-page":"593","article-title":"Asymmetric non-local neural networksfor semantic segmentation[C]\/\/","volume":"2019","author":"Zhu","journal-title":"Proceedings of the IEEEInternational Conference on Computer Vision"},{"key":"10.3233\/JIFS-210728_ref6","first-page":"1765","article-title":"Universaladversarial perturbations[C]\/\/","volume":"2017","author":"Moosavi-Dezfooli","journal-title":"Proceedings of the IEEEconference on computer vision and pattern recognition"},{"key":"10.3233\/JIFS-210728_ref9","first-page":"1000","article-title":"Adversarial camouflage: Hiding physical-world attacks with natural styles[C]\/\/","volume":"2020","author":"Duan","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition"},{"key":"10.3233\/JIFS-210728_ref10","first-page":"39","article-title":"Towards evaluating the robustness of neural networks[C]\/\/","volume":"2017","author":"Carlini","journal-title":"2017 ieee symposium on security and privacy (sp). IEEE"},{"key":"10.3233\/JIFS-210728_ref12","first-page":"1625","article-title":"Robust physical-worldattacks on deep learning visual classification[C]\/\/","volume":"2018","author":"Eykholt","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"10.3233\/JIFS-210728_ref13","first-page":"2574","article-title":"Deepfool: a simpleand accurate method to fool deep neural networks[C]\/\/","volume":"2016","author":"Moosavi-Dezfooli","journal-title":"Proceedings of the IEEE conference on computer vision and patternrecognition"},{"key":"10.3233\/JIFS-210728_ref14","first-page":"43","article-title":"Learning universal adversarial perturbations with generative models[C]\/\/","volume":"2018","author":"Hayes","journal-title":"2018 IEEE Security and Privacy Workshops (SPW). IEEE"},{"key":"10.3233\/JIFS-210728_ref15","first-page":"742","article-title":"NAG: Network for adversary generation[C]\/\/","volume":"2018","author":"Reddy Mopuri","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"10.3233\/JIFS-210728_ref16","first-page":"19","article-title":"Ask, acquire, and attack: Data-free uap generation using classimpressions[C]\/\/","volume":"2018","author":"Reddy Mopuri","journal-title":"Proceedings of the European Conference on Computer Vision (ECCV)"},{"issue":"10","key":"10.3233\/JIFS-210728_ref17","doi-asserted-by":"crossref","first-page":"2452","DOI":"10.1109\/TPAMI.2018.2861800","article-title":"Generalizable data-free objective for crafting universal adversarial perturbations[J]","volume":"41","author":"Mopuri","year":"2018","journal-title":"IEEE transactions on pattern analysis and machine intelligence"},{"key":"10.3233\/JIFS-210728_ref18","first-page":"2941","article-title":"Universal adversarial perturbation via prior driven uncertainty approximation[C]\/\/","volume":"2019","author":"Liu","journal-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision"},{"key":"10.3233\/JIFS-210728_ref19","first-page":"8562","article-title":"Art of singular vectors and universal adversarial perturbations[C]\/\/","volume":"2018","author":"Khrulkov","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"issue":"6","key":"10.3233\/JIFS-210728_ref20","doi-asserted-by":"crossref","first-page":"84","DOI":"10.1145\/3065386","article-title":"Imagenet classification with deep convolutional neural networks[J]","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Communications of the ACM"},{"key":"10.3233\/JIFS-210728_ref23","first-page":"1","article-title":"Going deeper with convolutions[C]\/\/","volume":"2015","author":"Szegedy","journal-title":"Proceedings of the IEEE conference on computer vision and pattern recognition"},{"key":"10.3233\/JIFS-210728_ref24","first-page":"770","article-title":"Deep residual learning for image recognition[C]\/\/","volume":"2016","author":"He","journal-title":"Proceedings of the IEEE conference on computer vision and pattern recognition"},{"issue":"3","key":"10.3233\/JIFS-210728_ref25","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","article-title":"Imagenet large scale visual recognition challenge[J]","volume":"115","author":"Russakovsky","year":"2015","journal-title":"International Journal of Computer Vision"},{"key":"10.3233\/JIFS-210728_ref27","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1016\/j.brainresbull.2020.06.007","article-title":"T-distribution stochastic neighbor embedding for fine brain functional parcellation on rs-fMRI[J]","volume":"162","author":"Hu","year":"2020","journal-title":"Brain Research Bulletin"},{"key":"10.3233\/JIFS-210728_ref28","first-page":"709","article-title":"Defending against universal attacks through selective feature regeneration[C]\/\/","volume":"2020","author":"Borkar","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition"}],"container-title":["Journal of Intelligent &amp; Fuzzy Systems"],"original-title":[],"link":[{"URL":"https:\/\/content.iospress.com\/download?id=10.3233\/JIFS-210728","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T09:45:23Z","timestamp":1777455923000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/full\/10.3233\/JIFS-210728"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,4,28]]},"references-count":19,"journal-issue":{"issue":"6"},"URL":"https:\/\/doi.org\/10.3233\/jifs-210728","relation":{},"ISSN":["1064-1246","1875-8967"],"issn-type":[{"value":"1064-1246","type":"print"},{"value":"1875-8967","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,4,28]]}}}