Recent Advances in Adversarial Training for Adversarial Robustness

Abstract

Adversarial training is one of the most effective approaches defending against adversarial examples for deep learning models. Unlike other defense strategies, adversarial training aims to promote the robustness of models intrinsically. During the last few years, adversarial training has been studied and discussed from various aspects. A variety of improvements and developments of adversarial training are proposed, which were, however, neglected in existing surveys. For the first time in this survey, we systematically review the recent progress on adversarial training for adversarial robustness with a novel taxonomy. Then we discuss the generalization problems in adversarial training from three perspectives. Finally, we highlight the challenges which are not fully tackled and present potential future directions.

accepted by International Joint Conference on Artificial Intelligence (IJCAI-21)

References in corpus (15)

• Improved Regularization of Convolutional Neural Networks with Cutout
• Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
• The Space of Transferable Adversarial Examples
• Adversarial Machine Learning at Scale
• Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients
• Improving Adversarial Robustness via Promoting Ensemble Diversity
• Improving Adversarial Robustness of Ensembles with Diversity Training
• Instance adaptive adversarial training: Improved accuracy tradeoffs in neural nets
• Metric Learning for Adversarial Robustness
• Adversarial Examples in Modern Machine Learning: A Review
• Robustness to Adversarial Perturbations in Learning from Incomplete Data
• Opportunities and Challenges in Deep Learning Adversarial Robustness: A Survey
• Transfer of Adversarial Robustness Between Perturbation Types
• Towards Understanding Fast Adversarial Training
• Regularizers for Single-step Adversarial Training