摘要
Constructing adversarial examples in a black-box threat model injures the original images by introducing visual distortion. In this paper, we propose a novel black-box attack approach that can directly minimize the induced distortion by learning the noise distribution of the adversarial example, assuming only loss-oracle access to the black-box network. To quantify visual distortion, the perceptual distance between the adversarial example and the original image, is introduced in our loss. We first approximate the gradient of the corresponding non-differentiable loss function by sampling noise from the learned noise distribution. Then the distribution is updated using the estimated gradient to reduce visual distortion. The learning continues until an adversarial example is found. We validate the effectiveness of our attack on ImageNet. Our attack results in much lower distortion when compared to the state-of-the-art black-box attacks and achieves 100% success rate on InceptionV3, ResNet50 and VGG16bn. Furthermore, we theoretically prove the convergence of our model. The code is publicly available at https://github.com/Alina-1997/visual-distortion-in-attack.
| 源语言 | 英语 |
|---|---|
| 页(从-至) | 6156-6167 |
| 页数 | 12 |
| 期刊 | IEEE Transactions on Image Processing |
| 卷 | 30 |
| DOI | |
| 出版状态 | 已出版 - 1月 2021 |
基金资助
This work was supported in part by the National Natural Science Foundation of China under Grant 62036005 and in part by the Fundamental Research Funds for the Central Universities under Grant 2042020KF0205.
关键词
- Distortion
- Visualization
- Measurement
- Loss measurement
- Optimization
- Convergence
- Training
- Black-box attack
- adversarial examples
- classification
收录
- SCIE
- EI
- Scopus
学术指纹
探究 'Toward Visual Distortion in Black-Box Attacks' 的科研主题。它们共同构成独一无二的学术指纹。引用此
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver