Decision-Based Adversarial Attack With Frequency Mixup

It has been widely observed that deep neural networks are highly vulnerable to adversarial examples. Decision-based attacks could generate adversarial examples based solely on top-1 labels returned by the target model. However, they typically make excessive queries and could not bypass detection eff...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on information forensics and security 2022, Vol.17, p.1038-1052
Main Authors: Li, Xiu-Chuan, Zhang, Xu-Yao, Yin, Fei, Liu, Cheng-Lin
Format: Article
Language:English
Subjects:
Artificial neural networks
Convolutional neural networks
Data models
Decision-based attack
detection
Distortion
frequency domain
Numerical models
Perturbation methods
Predictive models
Queries
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:It has been widely observed that deep neural networks are highly vulnerable to adversarial examples. Decision-based attacks could generate adversarial examples based solely on top-1 labels returned by the target model. However, they typically make excessive queries and could not bypass detection effectively. To comprehensively assess a decision-based attack, besides its query efficiency, the performance against detection is also a concern. Considering that previous detections consume massive resources and always mistakenly recognize benign video frames as malicious attacks, we design a lightweight detection called boundary detection to overcome the above limitations, whose success reveals serious limitations of existing decision-based attacks. To develop more powerful attacks, we first present f-mixup as a basic method to produce candidate adversarial examples in the frequency domain. Using f-mixup as the building block, we propose f-attack as a complete decision-based attack. With the help of several natural images, f-attack could both work well with limited (hundreds of) queries and bypass detection effectively. Nevertheless, if the attacker could make relatively adequate (thousands of) queries and the target model is not equipped with detection, f-attack will lag behind existing decision-based attacks. We additionally introduce frequency binary search based on f-mixup , which serves as a plug-and-play module for existing decision-based attacks to further improve their query efficiency. Experimental results verify the effectiveness of our proposed methods.
ISSN:1556-6013
1556-6021
DOI:10.1109/TIFS.2022.3156809