Benchmarking Adversarial Robustness

Yinpeng Dong, Qi-An Fu, Xiao Yang, Tianyu Pang, Hang Su, Jun Zhu

Keywords: Adversarial Example, Robustness, Benchmark

TL;DR: We provide a comprehensive, rigorous, and coherent benchmark to evaluate adversarial robustness of deep learning models.

Abstract: Deep neural networks are vulnerable to adversarial examples, which becomes one of the most important problems in the development of deep learning. While a lot of efforts have been made in recent years, it is of great significance to perform correct and complete evaluations of the adversarial attack and defense algorithms. In this paper, we establish a comprehensive, rigorous, and coherent benchmark to evaluate adversarial robustness on image classification tasks. After briefly reviewing plenty of representative attack and defense methods, we perform large-scale experiments with two robustness curves as the fair-minded evaluation criteria to fully understand the performance of these methods. Based on the evaluation results, we draw several important findings and provide insights for future research.

Code: https://github.com/adversarial-robustness-benchmark/adversarial-robustness-benchmark