Self-supervised Adversarial Training

Abstract

Recent work has demonstrated that neural networks are vulnerable to adversarial examples. To escape from the predicament, many works try to harden the model in various ways, in which adversarial training is an effective way which learns robust feature representation so as to resist adversarial attacks. Meanwhile, the self-supervised learning aims to learn robust and semantic embedding from data itself. With these views, we introduce self-supervised learning to against adversarial examples in this paper. Specifically, the self-supervised representation coupled with k-Nearest Neighbour is proposed for classification. To further strengthen the defense ability, self-supervised adversarial training is proposed, which maximizes the mutual information between the representations of original examples and the corresponding adversarial examples. Experimental results show that the self-supervised representation outperforms its supervised version in respect of robustness and self-supervised adversarial training can further improve the defense ability efficiently.

arXiv commentAccepted to ICASSP 2020

References in corpus (8)

• Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks
• Towards Deep Learning Models Resistant to Adversarial Attacks
• Learning Representations by Maximizing Mutual Information Across Views
• Adversarial Examples Are Not Bugs, They Are Features
• PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
• On Mutual Information Maximization for Representation Learning
• Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination
• Defending Against Adversarial Examples with K-Nearest Neighbor

Cited by in corpus (2)

• Adversarial Self-Supervised Contrastive Learning
• Language Guided Adversarial Purification