A3T: Adversarially Augmented Adversarial Training

Research paper by Akram Erraqabi, Aristide Baratin, Yoshua Bengio, Simon Lacoste-Julien

Indexed on: 11 Jan '18Published on: 11 Jan '18Published in: arXiv - Computer Science - Learning


Recent research showed that deep neural networks are highly sensitive to so-called adversarial perturbations, which are tiny perturbations of the input data purposely designed to fool a machine learning classifier. Most classification models, including deep learning models, are highly vulnerable to adversarial attacks. In this work, we investigate a procedure to improve adversarial robustness of deep neural networks through enforcing representation invariance. The idea is to train the classifier jointly with a discriminator attached to one of its hidden layer and trained to filter the adversarial noise. We perform preliminary experiments to test the viability of the approach and to compare it to other standard adversarial training methods.