EUVIMEDEUVIMED entdecken
Crossref

Boosting Adversarial Transferability across Model Genus by Deformation-Constrained Warping

Qinliang Lin, Cheng Luo, Zenghao Niu, Xilin He, Weicheng Xie, Yuanbo Hou, Linlin Shen, Siyang Song

Proceedings of the AAAI Conference on Artificial Intelligence · 24.03.2024 · 38 (4) · 3459-3467

DOI: 10.1609/aaai.v38i4.28133 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 24.03.2024. Verfasst von Qinliang Lin und 7 weiteren Autor:innen. Die Arbeit wurde bisher 28-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Adversarial examples generated by a surrogate model typically exhibit limited transferability to unknown target systems. To address this problem, many transferability enhancement approaches (e.g., input transformation and model augmentation) have been proposed. However, they show poor performances in attacking systems having different model genera from the surrogate model. In this paper, we propose a novel and generic attacking strategy, called Deformation-Constrained Warping Attack (DeCoWA), that can be effectively applied to cross model genus attack. Specifically, DeCoWA firstly augments input examples via an elastic deformation, namely Deformation-Constrained Warping (DeCoW), to obtain rich local details of the augmented input. To avoid severe distortion of global semantics led by random deformation, DeCoW further constrains the strength and direction of the warping transformation by a novel adaptive control strategy. Extensive experiments demonstrate that the transferable examples crafted by our DeCoWA on CNN surrogates can significantly hinder the performance of Transformers (and vice versa) on various tasks, including image classification, video action recognition, and audio recognition. Code is made available at https://github.com/LinQinLiang/DeCoWA.

Zitieren

APA

Lin, Q., Luo, C., Niu, Z., He, X., Xie, W., Hou, Y., Shen, L., & Song, S. (2024). Boosting Adversarial Transferability across Model Genus by Deformation-Constrained Warping. Proceedings of the AAAI Conference on Artificial Intelligence, 38(4), 3459-3467. https://doi.org/10.1609/aaai.v38i4.28133

Vancouver

Lin Q, Luo C, Niu Z, He X, Xie W, Hou Y, et al. Boosting Adversarial Transferability across Model Genus by Deformation-Constrained Warping. Proceedings of the AAAI Conference on Artificial Intelligence. 2024;38(4):3459-3467. doi:10.1609/aaai.v38i4.28133

BibTeX
@article{Lin2024,
  author = {Lin, Qinliang and Luo, Cheng and Niu, Zenghao and He, Xilin and Xie, Weicheng and Hou, Yuanbo and Shen, Linlin and Song, Siyang},
  title = {Boosting Adversarial Transferability across Model Genus by Deformation-Constrained Warping},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2024},
  volume = {38},
  number = {4},
  pages = {3459--3467},
  doi = {10.1609/aaai.v38i4.28133}
}
Kostenlos starten →