Saved in:
Bibliographic Details
Main Authors: Zheng, Yayin, Wan, Chen, Guo, Zihong, Kuang, Hailing, Lu, Xiaohai
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.21181
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912398082834432
author Zheng, Yayin
Wan, Chen
Guo, Zihong
Kuang, Hailing
Lu, Xiaohai
author_facet Zheng, Yayin
Wan, Chen
Guo, Zihong
Kuang, Hailing
Lu, Xiaohai
contents Adversarial attacks have become a significant challenge in the security of machine learning models, particularly in the context of black-box defense strategies. Existing methods for enhancing adversarial transferability primarily focus on the spatial domain. This paper presents Frequency-Space Attack (FSA), a new adversarial attack framework that effectively integrates frequency-domain and spatial-domain transformations. FSA combines two key techniques: (1) High-Frequency Augmentation, which applies Fourier transform with frequency-selective amplification to diversify inputs and emphasize the critical role of high-frequency components in adversarial attacks, and (2) Hierarchical-Gradient Fusion, which merges multi-scale gradient decomposition and fusion to capture both global structures and fine-grained details, resulting in smoother perturbations. Our experiment demonstrates that FSA consistently outperforms state-of-the-art methods across various black-box models. Notably, our proposed FSA achieves an average attack success rate increase of 23.6% compared with BSR (CVPR 2024) on eight black-box defense models.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21181
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Boosting Adversarial Transferability via High-Frequency Augmentation and Hierarchical-Gradient Fusion
Zheng, Yayin
Wan, Chen
Guo, Zihong
Kuang, Hailing
Lu, Xiaohai
Computer Vision and Pattern Recognition
Image and Video Processing
Adversarial attacks have become a significant challenge in the security of machine learning models, particularly in the context of black-box defense strategies. Existing methods for enhancing adversarial transferability primarily focus on the spatial domain. This paper presents Frequency-Space Attack (FSA), a new adversarial attack framework that effectively integrates frequency-domain and spatial-domain transformations. FSA combines two key techniques: (1) High-Frequency Augmentation, which applies Fourier transform with frequency-selective amplification to diversify inputs and emphasize the critical role of high-frequency components in adversarial attacks, and (2) Hierarchical-Gradient Fusion, which merges multi-scale gradient decomposition and fusion to capture both global structures and fine-grained details, resulting in smoother perturbations. Our experiment demonstrates that FSA consistently outperforms state-of-the-art methods across various black-box models. Notably, our proposed FSA achieves an average attack success rate increase of 23.6% compared with BSR (CVPR 2024) on eight black-box defense models.
title Boosting Adversarial Transferability via High-Frequency Augmentation and Hierarchical-Gradient Fusion
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2505.21181