On Detecting Adversarial Perturbations
This addresses the security problem of adversarial attacks in machine learning for practitioners, offering a novel detection approach rather than robustness enhancement.
The paper tackles the vulnerability of deep neural networks to adversarial perturbations by proposing a detector subnetwork to distinguish genuine from adversarial data, showing that such perturbations can be detected well and that detectors generalize to similar adversaries.
Machine learning and deep learning in particular has advanced tremendously on perceptual tasks in recent years. However, it remains vulnerable against adversarial perturbations of the input that have been crafted specifically to fool the system while being quasi-imperceptible to a human. In this work, we propose to augment deep neural networks with a small "detector" subnetwork which is trained on the binary classification task of distinguishing genuine data from data containing adversarial perturbations. Our method is orthogonal to prior work on addressing adversarial perturbations, which has mostly focused on making the classification network itself more robust. We show empirically that adversarial perturbations can be detected surprisingly well even though they are quasi-imperceptible to humans. Moreover, while the detectors have been trained to detect only a specific adversary, they generalize to similar and weaker adversaries. In addition, we propose an adversarial attack that fools both the classifier and the detector and a novel training procedure for the detector that counteracts this attack.