English

Attacking the Madry Defense Model with $L_1$-based Adversarial Examples

Machine Learning 2018-07-31 v4 Cryptography and Security Machine Learning

Abstract

The Madry Lab recently hosted a competition designed to test the robustness of their adversarially trained MNIST model. Attacks were constrained to perturb each pixel of the input image by a scaled maximal LL_\infty distortion ϵ\epsilon = 0.3. This discourages the use of attacks which are not optimized on the LL_\infty distortion metric. Our experimental results demonstrate that by relaxing the LL_\infty constraint of the competition, the elastic-net attack to deep neural networks (EAD) can generate transferable adversarial examples which, despite their high average LL_\infty distortion, have minimal visual distortion. These results call into question the use of LL_\infty as a sole measure for visual distortion, and further demonstrate the power of EAD at generating robust adversarial examples.

Keywords

Cite

@article{arxiv.1710.10733,
  title  = {Attacking the Madry Defense Model with $L_1$-based Adversarial Examples},
  author = {Yash Sharma and Pin-Yu Chen},
  journal= {arXiv preprint arXiv:1710.10733},
  year   = {2018}
}

Comments

Accepted to ICLR 2018 Workshops

R2 v1 2026-06-22T22:29:11.269Z