Rethinking Model Ensemble in Transfer-based Adversarial Attacks

Huanran Chen, Yichi Zhang, Yinpeng Dong, Xiao Yang, Hang Su, Jun Zhu

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: Adversarial attack, transferability, ensemble attack, robustness

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: We rethink model ensemble in transfer-based adversarial attacks and propose a Common Weakness Attack to generate more transferable adversarial examples.

Abstract: It is widely recognized that deep learning models lack robustness to adversarial examples. An intriguing property of adversarial examples is that they can transfer across different models, which enables black-box attacks without any knowledge of the victim model. An effective strategy to improve the transferability is attacking an ensemble of models. However, previous works simply average the outputs of different models, lacking an in-depth analysis on how and why model ensemble methods can strongly improve the transferability. In this paper, we rethink the ensemble in adversarial attacks and define the common weakness of model ensemble with two properties: 1) the flatness of loss landscape; and 2) the closeness to the local optimum of each model. We empirically and theoretically show that both properties are strongly correlated with the transferability and propose a Common Weakness Attack (CWA) to generate more transferable adversarial examples by promoting these two properties. Experimental results on both image classification and object detection tasks validate the effectiveness of our approach to improving the adversarial transferability, especially when attacking adversarially trained models. We also successfully apply our method to attack a black-box large vision-language model -- Google's Bard, showing the practical effectiveness. Code is available at \url{https://github.com/huanranchen/AdversarialAttacks}.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Primary Area: societal considerations including fairness, safety, privacy

Submission Number: 30