MGAAttack: Toward More Query-efficient Black-box Attack by Microbial Genetic Algorithm

Lina Wang, Kang Yang, Wenqi Wang, Run Wang, Aoshuang Ye

Abstract

Recent studies have shown that deep neural networks (DNNs) are susceptible to adversarial attacks even in the black-box settings. However, previous studies on creating black-box based adversarial examples by merely solving the traditional continuous problem, which suffer query efficiency issues. To address the efficiency of querying in black-box attack, we propose a novel attack, called MGAAttack, which is a query-efficient and gradient-free black-box attack without obtaining any knowledge of the target model. In our approach, we leverage the advantages of both transfer-based and scored-based methods, two typical techniques in black-box attack, and solve a discretized problem by using a simple yet effective microbial genetic algorithm (MGA). Experimental results show that our approach dramatically reduces the number of queries on CIFAR-10 and ImageNet and significantly outperforms previous work. In the untargeted attack, we can attack a VGG19 classifier with only 16 queries and give an attack success rate more than 99.90% on ImageNet. Our code is available at https://github.com/kangyangWHU/MGAAttack.
BibTeX
@inproceedings{wang2020mgaattack,
  author = {Lina Wang and Kang Yang and Wenqi Wang and Run Wang and Aoshuang Ye},
  title = {{MGAAttack: Toward More Query-efficient Black-box Attack by Microbial Genetic Algorithm}},
  booktitle = {ACM Multimedia},
  year = {2020},
  doi = {10.1145/3394171.3413703},
}

Loading related content...

Figure