Visual Interpretability Alone Helps Adversarial Robustness

25 Sep 2019  ·  Akhilan Boopathy, Sijia Liu, Gaoyuan Zhang, Pin-Yu Chen, Shiyu Chang, Luca Daniel ·

Recent works have empirically shown that there exist adversarial examples that can be hidden from neural network interpretability, and interpretability is itself susceptible to adversarial attacks. In this paper, we theoretically show that with the correct measurement of interpretation, it is actually difficult to hide adversarial examples, as confirmed by experiments on MNIST, CIFAR-10 and Restricted ImageNet. Spurred by that, we develop a novel defensive scheme built only on robust interpretation (without resorting to adversarial loss minimization). We show that our defense achieves similar classification robustness to state-of-the-art robust training methods while attaining higher interpretation robustness under various settings of adversarial attacks.

PDF Abstract
No code implementations yet. Submit your code now


  Add Datasets introduced or used in this paper

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.


No methods listed for this paper. Add relevant methods here