paper-with-me

Papers

Query-Efficient Decision-based Black-Box Patch Attack

2023-07-02 · Zhaoyu Chen, Bo Li, Shuang Wu, Shouhong Ding, Wenqiang Zhang

Deep neural networks (DNNs) have been showed to be highly vulnerable to imperceptible adversarial perturbations. As a complementary type of adversary, patch attacks that introduce perceptible perturbations to the images have attracted the interest of researchers. Existing patch attacks rely on the architecture of the model or the probabilities of predictions and perform poorly in the decision-based setting, which can still construct a perturbation with the minimal information exposed -- the top-1 predicted label. In this work, we first explore the decision-based patch attack. To enhance the attack efficiency, we model the patches using paired key-points and use targeted images as the initialization of patches, and parameter optimizations are all performed on the integer domain. Then, we propose a differential evolutionary algorithm named DevoPatch for query-efficient decision-based patch attacks. Experiments demonstrate that DevoPatch outperforms the state-of-the-art black-box patch attacks in terms of patch area and attack success rate within a given query budget on image classification and face verification. Additionally, we conduct the vulnerability evaluation of ViT and MLP on image classification in the decision-based patch attack setting for the first time. Using DevoPatch, we can evaluate the robustness of models to black-box patch attacks. We believe this method could inspire the design and deployment of robust vision models based on various DNN architectures in the future.

📄 PDF Abstract BibTeX arXiv:2307.00477

Code (0)

등록된 구현이 없습니다.

Tasks

Face Verificationimage-classificationImage Classification

Similar Papers 제목 키워드 기반

Decision-based Black-box Attack Against Vision Transformers via Patch-wise Adversarial Removal

2021-12-07 · Yucheng Shi, Yahong Han, Yu-an Tan, Xiaohui Kuang

Vision transformers (ViTs) have demonstrated impressive performance and stronger adversarial robustness compared to Convolutional Neural Networks (CNNs). On the one hand, ViTs' focus on global interaction between individ…

Adversarial RobustnessSensitivity

Efficient Decision-based Black-box Patch Attacks on Video Recognition

2023-03-21 · ICCV 2023 1 · Kaixun Jiang, Zhaoyu Chen, Hao Huang, Jiafeng Wang 외

Although Deep Neural Networks (DNNs) have demonstrated excellent performance, they are vulnerable to adversarial patches that introduce perceptible and localized perturbations to the input. Generating adversarial patches…

Video Recognition

Query-Efficient Hard-Label Black-Box Attack against Vision Transformers

2024-06-29 · Chao Zhou, Xiaowen Shi, Yuan-Gen Wang

Recent studies have revealed that vision transformers (ViTs) face similar security risks from adversarial attacks as deep convolutional neural networks (CNNs). However, directly applying attack methodology on CNNs to ViT…

Adversarial Attack

Sparse-RS: a versatile framework for query-efficient sparse black-box adversarial attacks

2020-06-23 · Francesco Croce, Maksym Andriushchenko, Naman D. Singh, Nicolas Flammarion 외

We propose a versatile framework based on random search, Sparse-RS, for score-based sparse targeted and untargeted attacks in the black-box setting. Sparse-RS does not rely on substitute models and achieves state-of-the-…

Malware Detection

PatchAttack: A Black-box Texture-based Attack with Reinforcement Learning

2020-04-12 · ECCV 2020 8 · Chenglin Yang, Adam Kortylewski, Cihang Xie, Yinzhi Cao 외

Patch-based attacks introduce a perceptible but localized change to the input that induces misclassification. A limitation of current patch-based black-box attacks is that they perform poorly for targeted attacks, and ev…

Adversarial DefenseClusteringreinforcement-learningReinforcement Learning+1