Towards Speeding up Adversarial Training in Latent Spaces

TL;DR

This paper introduces a novel adversarial training method that generates adversarial examples in the latent space, significantly reducing training time while improving robustness and maintaining accuracy.

Contribution

The paper proposes Endogenous Adversarial Examples (EAEs) in latent space, eliminating the need for costly adversarial example generation in input space.

Findings

Reduces adversarial training time on CIFAR-10 and ImageNet

Enhances model robustness against adversarial attacks

Maintains higher accuracy on clean data compared to existing methods

Abstract

Adversarial training is wildly considered as one of the most effective way to defend against adversarial examples. However, existing adversarial training methods consume unbearable time, due to the fact that they need to generate adversarial examples in the large input space. To speed up adversarial training, we propose a novel adversarial training method that does not need to generate real adversarial examples. By adding perturbations to logits to generate Endogenous Adversarial Examples (EAEs) -- the adversarial examples in the latent space, the time consuming gradient calculation can be avoided. Extensive experiments are conducted on CIFAR-10 and ImageNet, and the results show that comparing to state-of-the-art methods, our EAE adversarial training not only shortens the training time, but also enhances the robustness of the model and has less impact on the accuracy of clean examples than the existing methods.