Timezone: »

Fast and Reliable Evaluation of Adversarial Robustness with Minimum-Margin Attack
Ruize Gao · Jiongxiao Wang · Kaiwen Zhou · Feng Liu · Binghui Xie · Gang Niu · Bo Han · James Cheng

Tue Jul 19 01:50 PM -- 01:55 PM (PDT) @ None

The AutoAttack (AA) has been the most reliable method to evaluate adversarial robustness when considerable computational resources are available. However, the high computational cost (e.g., 100 times more than that of the project gradient descent attack) makes AA infeasible for practitioners with limited computational resources, and also hinders applications of AA in the adversarial training (AT). In this paper, we propose a novel method, minimum-margin (MM) attack, to fast and reliably evaluate adversarial robustness. Compared with AA, our method achieves comparable performance but only costs 3% of the computational time in extensive experiments. The reliability of our method lies in that we evaluate the quality of adversarial examples using the margin between two targets that can precisely identify the most adversarial example. The computational efficiency of our method lies in an effective Sequential TArget Ranking Selection (STARS) method, ensuring that the cost of the MM attack is independent of the number of classes. The MM attack opens a new way for evaluating adversarial robustness and provides a feasible and reliable way to generate high-quality adversarial examples in AT.

Author Information

Ruize Gao (The Chinese University of Hong Kong)
Jiongxiao Wang (Fudan University)
Kaiwen Zhou (The Chinese University of Hong Kong)
Feng Liu (The University of Melbourne)
Binghui Xie (The Chinese University of Hong Kong)
Gang Niu (RIKEN)
Gang Niu

Gang Niu is currently an indefinite-term research scientist at RIKEN Center for Advanced Intelligence Project.

James Cheng (CUHK)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors