Improving Transferability of Adversarial Attacks via Frequency-Consistent Regularization

Tengfei Shi

doi:10.3390/app16083748

2026-04-11

DOI：10.3390/app16083748

原文链接

原文求助

摘要

En 中文

对抗样本揭示了深度神经网络的脆弱性，其迁移性使得黑盒攻击尤为令人担忧。然而，在代理模型上精心制作的扰动往往在未见过的目标模型上效果不足。本文从频域视角重新审视这一问题，并观察到扰动优化可能过度依赖特定的频谱模式，从而削弱跨模型迁移能力。为解决该问题，我们提出频谱一致性正则化（FCR），一种简单的插件策略，可与现有迭代攻击结合使用。FCR在每个迭代中引入多个保留低频特征的视图，并随机采样频率范围，随后在这些不同视图中优化扰动。通过这种方式，生成的扰动对特定频谱配置的依赖性降低，表现出更优的迁移性。实验结果表明，FCR能显著提升各类迭代攻击的迁移性能。该改进不仅体现在标准目标模型上，在对抗训练模型中效果更为显著。

Keyword:

adversarial examples

deep neural networks

black-box adversarial attack

transferable attack

iterative attacks

transferability

期刊

Applied Sciences-Basel

IF:

2.5

论文数:

7.6K

被引数:

机构

Beihang University

学者数:

5.2W

论文数: 4.1W

被引数: 37

引用论文

Powerful Physical Adversarial Examples Against Practical Face Recognition Systems

2022 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW)

IF0

2022-01-01

OAAI

Inderjeet Singh; Toshinori Araki; Kazuya Kakizaki

Image Super-Resolution as a Defense Against Adversarial Attacks

IEEE TRANSACTIONS ON IMAGE PROCESSING

IF13.7

2020-01-01

OAAI

Mustafa, Aamir; Khan, Salman H.; Hayat, Munawar; Shen, Jianbing; Shao, Ling

Rethinking the Inception Architecture for Computer Vision 重新思考计算机视觉的Inception体系结构

2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

IF0

2016-06-01

OAAI

Christian Szegedy; Vincent Vanhoucke; Sergey Ioffe; Jon Shlens; Zbigniew Wojna

Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning Inception-v4、Inception-ResNet和剩余连接对学习的影响

Proceedings of the AAAI Conference on Artificial Intelligence

IF0

2017-02-12

OAAI

Christian Szegedy; Sergey Ioffe; Vincent Vanhoucke; Alexander Alemi

Feature Distillation: DNN-Oriented JPEG Compression Against Adversarial Examples

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

IF0

2019-06-01

OAAI

Zihao Liu; Qi Liu; Tao Liu; Nuo Xu; Xue Lin; Yanzhi Wang; Wujie Wen

Structure Invariant Transformation for better Adversarial Transferability 为更好的对抗可转移性而进行的结构不变变换

2023 IEEE/CVF International Conference on Computer Vision (ICCV)

IF0

2023-10-01

OAAI

Xiaosen Wang; Zeliang Zhang; Jianping Zhang

Deflecting Adversarial Attacks with Pixel Deflection

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

IF0

2018-06-01

OAAI

Aaditya Prakash; Nick Moran; Solomon Garber; Antonella DiLillo; James Storer

Boosting Adversarial Attacks with Momentum 用动量增强对抗性攻击

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

IF0

2018-06-01

OAAI

Yinpeng Dong; Fangzhou Liao; Tianyu Pang; Hang Su; Jun Zhu; Xiaolin Hu; Jianguo Li

Studying the robustness of data imputation methodologies against adversarial attacks 研究数据插补方法对抗攻击的鲁棒性

Computers & Security

IF5.4

2025-06-30

PREAI

Arthur Dantas Mangussi; Ricardo Cardoso Pereira; Ana Carolina Lorena; Miriam Seoane Santos; Pedro Henriques Abreu

ImageNet Large Scale Visual Recognition Challenge ImageNet大规模视觉识别挑战

INTERNATIONAL JOURNAL OF COMPUTER VISION

IF9.3

2015-04-11

2.7W

PREAI

Russakovsky, Olga; Deng, Jia; Su, Hao; Krause, Jonathan; Satheesh, Sanjeev; Ma, Sean; Huang, Zhiheng; Karpathy, Andrej; Khosla, Aditya; Bernstein, Michael; Berg, Alexander C.; Fei-Fei, Li

查看更多内容