paper-with-me

홈 › Papers

Rethinking Model Ensemble in Transfer-based Adversarial Attacks

2023-03-16 · Huanran Chen, Yichi Zhang, Yinpeng Dong, Xiao Yang, Hang Su, Jun Zhu

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}.

📄 PDF Abstract BibTeX arXiv:2303.09105

Code (2)

huanranchen/AdversarialAttacks 공식 구현 pytorch
huanranchen/CommonWeaknessDetection pytorch

Tasks

image-classificationImage ClassificationLanguage Modellingmodelobject-detectionObject Detection

Similar Papers 제목 키워드 기반

An Adaptive Model Ensemble Adversarial Attack for Boosting Adversarial Transferability

2023-08-05 · ICCV 2023 1 · Bin Chen, Jia-Li Yin, Shukai Chen, Bo-Hao Chen 외

While the transferability property of adversarial examples allows the adversary to perform black-box attacks (i.e., the attacker has no knowledge about the target model), the transfer-based adversarial attacks have gaine…

Adversarial Attack

ViT-EnsembleAttack: Augmenting Ensemble Models for Stronger Adversarial Transferability in Vision Transformers

2025-08-17 · Hanwen Cao, Haobo Lu, Xiaosen Wang, Kun He arxiv

Ensemble-based attacks have been proven to be effective in enhancing adversarial transferability by aggregating the outputs of models with various architectures. However, existing research primarily focuses on refining e…

Improving Adversarial Robustness of Ensembles with Diversity Training

2019-01-28 · Sanjay Kariyappa, Moinuddin K. Qureshi

Deep Neural Networks are vulnerable to adversarial attacks even in settings where the attacker has no direct access to the model being attacked. Such attacks usually rely on the principle of transferability, whereby an a…

Adversarial RobustnessDiversity

Guardian of the Ensembles: Introducing Pairwise Adversarially Robust Loss for Resisting Adversarial Attacks in DNN Ensembles

2021-12-09 · Shubhi Shukla, Subhadeep Dalui, Manaar Alam, Shubhajit Datta 외

Adversarial attacks rely on transferability, where an adversarial example (AE) crafted on a surrogate classifier tends to mislead a target classifier. Recent ensemble methods demonstrate that AEs are less likely to misle…

DiversityImage Classification

Ensemble Diversity Facilitates Adversarial Transferability

2024-01-01 · CVPR 2024 1 · Bowen Tang, Zheng Wang, Yi Bin, Qi Dou 외

With the advent of ensemble-based attacks the transferability of generated adversarial examples is elevated by a noticeable margin despite many methods only employing superficial integration yet ignoring the diversit…

Diversityreinforcement-learningReinforcement Learning