DC Field | Value | Language |
---|---|---|
dc.contributor.author | Zhang, Chaoning | ko |
dc.contributor.author | Zhang, Kang | ko |
dc.contributor.author | Zhang, Chenshuang | ko |
dc.contributor.author | Niu, Axi | ko |
dc.contributor.author | Feng, Jiu | ko |
dc.contributor.author | Yoo, Chang D. | ko |
dc.contributor.author | Kweon, In-So | ko |
dc.date.accessioned | 2022-11-15T09:00:22Z | - |
dc.date.available | 2022-11-15T09:00:22Z | - |
dc.date.created | 2022-11-15 | - |
dc.date.created | 2022-11-15 | - |
dc.date.issued | 2022-10 | - |
dc.identifier.citation | European Conference on Computer Vision, ECCV 2022, pp.725 - 742 | - |
dc.identifier.issn | 0302-9743 | - |
dc.identifier.uri | http://hdl.handle.net/10203/299664 | - |
dc.description.abstract | Adversarial training (AT) for robust representation learning and self-supervised learning (SSL) for unsupervised representation learning are two active research fields. Integrating AT into SSL, multiple prior works have accomplished a highly significant yet challenging task: learning robust representation without labels. A widely used framework is adversarial contrastive learning which couples AT and SSL, and thus constitutes a very complex optimization problem. Inspired by the divide-and-conquer philosophy, we conjecture that it might be simplified as well as improved by solving two sub-problems: non-robust SSL and pseudo-supervised AT. This motivation shifts the focus of the task from seeking an optimal integrating strategy for a coupled problem to finding sub-solutions for sub-problems. With this said, this work discards prior practices of directly introducing AT to SSL frameworks and proposed a two-stage framework termed Decoupled Adversarial Contrastive Learning (DeACL). Extensive experimental results demonstrate that our DeACL achieves SOTA self-supervised adversarial robustness while significantly reducing the training time, which validates its effectiveness and efficiency. Moreover, our DeACL constitutes a more explainable solution, and its success also bridges the gap with semi-supervised AT for exploiting unlabeled samples for robust representation learning. The code is publicly accessible at https://github.com/pantheon5100/DeACL. | - |
dc.language | English | - |
dc.publisher | Springer Nature Switzerland | - |
dc.title | Decoupled Adversarial Contrastive Learning for Self-supervised Adversarial Robustness | - |
dc.type | Conference | - |
dc.identifier.wosid | 000903586400042 | - |
dc.identifier.scopusid | 2-s2.0-85144544122 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 725 | - |
dc.citation.endingpage | 742 | - |
dc.citation.publicationname | European Conference on Computer Vision, ECCV 2022 | - |
dc.identifier.conferencecountry | IS | - |
dc.identifier.conferencelocation | Tel Aviv | - |
dc.identifier.doi | 10.1007/978-3-031-20056-4_42 | - |
dc.contributor.localauthor | Yoo, Chang D. | - |
dc.contributor.localauthor | Kweon, In-So | - |
dc.contributor.nonIdAuthor | Zhang, Chaoning | - |
dc.contributor.nonIdAuthor | Zhang, Kang | - |
dc.contributor.nonIdAuthor | Zhang, Chenshuang | - |
dc.contributor.nonIdAuthor | Niu, Axi | - |
dc.contributor.nonIdAuthor | Feng, Jiu | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.