Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Eliminating Catastrophic Overfitting Via Abnormal Adversarial Examples Regularization

About

Single-step adversarial training (SSAT) has demonstrated the potential to achieve both efficiency and robustness. However, SSAT suffers from catastrophic overfitting (CO), a phenomenon that leads to a severely distorted classifier, making it vulnerable to multi-step adversarial attacks. In this work, we observe that some adversarial examples generated on the SSAT-trained network exhibit anomalous behaviour, that is, although these training samples are generated by the inner maximization process, their associated loss decreases instead, which we named abnormal adversarial examples (AAEs). Upon further analysis, we discover a close relationship between AAEs and classifier distortion, as both the number and outputs of AAEs undergo a significant variation with the onset of CO. Given this observation, we re-examine the SSAT process and uncover that before the occurrence of CO, the classifier already displayed a slight distortion, indicated by the presence of few AAEs. Furthermore, the classifier directly optimizing these AAEs will accelerate its distortion, and correspondingly, the variation of AAEs will sharply increase as a result. In such a vicious circle, the classifier rapidly becomes highly distorted and manifests as CO within a few iterations. These observations motivate us to eliminate CO by hindering the generation of AAEs. Specifically, we design a novel method, termed Abnormal Adversarial Examples Regularization (AAER), which explicitly regularizes the variation of AAEs to hinder the classifier from becoming distorted. Extensive experiments demonstrate that our method can effectively eliminate CO and further boost adversarial robustness with negligible additional computational overhead.

Runqi Lin, Chaojian Yu, Tongliang Liu• 2024

Related benchmarks

TaskDatasetResultRank
Image ClassificationImageNet-100--
163
Image ClassificationImageNet-100 (test)
Clean Accuracy48.26
123
Image ClassificationCIFAR-100
Clean Accuracy57.1
90
Image ClassificationCIFAR-10
Clean Accuracy79.42
89
Image ClassificationPathMNIST
Clean Accuracy84.2
60
Medical Image ClassificationPathMNIST
Clean Accuracy84.2
48
Image ClassificationCIFAR100
Robust Accuracy22.93
34
Image ClassificationSVHN
Accuracy (Natural)94.99
30
Image ClassificationTiny ImageNet (test)
Standard Accuracy49.86
30
Image ClassificationTinyImageNet
Clean Accuracy45.11
30
Showing 10 of 27 rows

Other info

Code

Follow for update