Prior-Guided Adversarial Initialization for Fast Adversarial Training
Xiaojun Jia, Yong Zhang, Xingxing Wei, Baoyuan Wu, Ke Ma, Jue Wang, Xiaochun Cao
"Fast adversarial training (FAT) eï¬€ectively improves the efficiency of standard adversarial training (SAT). However, initial FAT encounters catastrophic overfitting, i.e., the robust accuracy against adversarial attacks suddenly decreases to 0% during training. Though several FAT variants spare no eï¬€ort to prevent overfitting, they sacrifice much calculation cost. In this paper, we explore the diï¬€erence between the training processes of SAT and FAT and observe that the attack success rate of adversarial examples (AEs) of FAT gets worse gradually in the late training stage, resulting in overfitting. The AEs are generated by the fast gradient sign method (FGSM) with a zero or random initialization. Based on the observation, we propose a prior-guided FGSM initialization method to avoid overfitting after investigating several initialization strategies, improving the quality of the AEs during the whole training process. The initialization is formed by leveraging historically generated AEs without additional calculation cost. We further provide a theoretical analysis for the proposed initialization method. Moreover, we also propose a simple yet eï¬€ective regularizer based on the prior guided initialization, i.e., the currently generated perturbation should not deviate too much from the prior-guided initialization. The regularizer adopts both historical and current adversarial perturbations to guide the model learning. Evaluations on four datasets demonstrate that the proposed method can prevent catastrophic overfitting and outperform state-of-the-art FAT methods at a low computational cost."