Document Type
Conference Proceeding
Publication Title
Proceedings of Machine Learning Research
Abstract
Deep neural networks have been found to be vulnerable to adversarial noise. Recent works show that exploring the impact of adversarial noise on intrinsic components of data can help improve adversarial robustness. However, the pattern closely related to human perception has not been deeply studied. In this paper, inspired by the cognitive science, we investigate the interference of adversarial noise from the perspective of image phase, and find ordinarily-trained models lack enough robustness against phase-level perturbations. Motivated by this, we propose a joint adversarial defense method: a phase-level adversarial training mechanism to enhance the adversarial robustness on the phase pattern; an amplitude-based pre-processing operation to mitigate the adversarial perturbation in the amplitude pattern. Experimental results show that the proposed method can significantly improve the robust accuracy against multiple attacks and even adaptive attacks. In addition, ablation studies demonstrate the effectiveness of our defense strategy.
First Page
42724
Last Page
42741
Publication Date
7-2023
Keywords
Cognitive science, Defense strategy, Human perception, Phase levels, Phase patterns, Pre-processing operations
Recommended Citation
D. Zhou, D. Zhou, N. Wang, H. Yang, X. Gao and T. Liu, "Phase-aware Adversarial Defense for Improving Adversarial Robustness," Proceedings of Machine Learning Research, vol. 202, pp. 42724 - 42741, Jul 2023.
Comments
Open Access version from PMLR
Uploaded on June 13, 2024