Towards Deep Learning Models Resistant to Adversarial Attacks.

Aleksander Madry, Aleksandar Makelov, Ludwig Schmidt, Dimitris Tsipras, Adrian Vladu

Browse the full ICLR paper archive.