Learning with a Strong Adversary

Understand

The robustness of neural networks to intended perturbations has recently attracted significant attention.

• In this paper, we propose a new method, \emph{learning with a strong adversary}, that learns robust classifiers from supervised data.
• The proposed method takes finding adversarial examples as an intermediate step.
• A new and simple way of finding adversarial examples is presented and experimentally shown to be efficient.