Timezone: »

Learning Black-Box Attackers with Transferable Priors and Query Feedback
Jiancheng YANG · Yangzhou Jiang · Xiaoyang Huang · Bingbing Ni · Chenglong Zhao

Mon Dec 07 09:00 PM -- 11:00 PM (PST) @ Poster Session 0 #127

This paper addresses the challenging black-box adversarial attack problem, where only classification confidence of a victim model is available. Inspired by consistency of visual saliency between different vision models, a surrogate model is expected to improve the attack performance via transferability. By combining transferability-based and query-based black-box attack, we propose a surprisingly simple baseline approach (named SimBA++) using the surrogate model, which significantly outperforms several state-of-the-art methods. Moreover, to efficiently utilize the query feedback, we update the surrogate model in a novel learning scheme, named High-Order Gradient Approximation (HOGA). By constructing a high-order gradient computation graph, we update the surrogate model to approximate the victim model in both forward and backward pass. The SimBA++ and HOGA result in Learnable Black-Box Attack (LeBA), which surpasses previous state of the art by considerable margins: the proposed LeBA significantly reduces queries, while keeping higher attack success rates close to 100% in extensive ImageNet experiments, including attacking vision benchmarks and defensive models. Code is open source at https://github.com/TrustworthyDL/LeBA.

Author Information

Jiancheng YANG (Shanghai Jiao Tong University)
Yangzhou Jiang (Shanghai Jiaotong University)

I'm a graduate student from Shanghai JIaotong University. My research interests lie on CV & ML security

Xiaoyang Huang (Shanghai Jiao Tong University)
Bingbing Ni (Shanghai Jiao Tong University)
Chenglong Zhao (Shanghai Jiao Tong University)

More from the Same Authors