Optimization for Robustness Evaluation Beyond ℓp Metrics

• Published in: ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 1 - 5
• Main Authors: Liang, Hengyue, Liang, Buyun, Cui, Ying, Mitchell, Tim, Sun, Ju
• Format: Conference Proceeding
• Language: English
• Published: IEEE 04.06.2023
• Subjects: adversarial attack; adversarial robustness; constrained optimization; Deep learning; deep neural networks; Measurement; Numerical models; perceptual distance; Perturbation methods; Robustness; robustness evaluation; Signal processing; Signal processing algorithms
• ISSN: 2379-190X
• DOI: 10.1109/ICASSP49357.2023.10095871

Empirical evaluation of the adversarial robustness of deep learning models involves solving non-trivial constrained optimization problems. Popular numerical algorithms to solve these constrained problems rely predominantly on projected gradient descent (PGD) and mostly handle adversarial perturbations modeled by the ℓ 1 , ℓ 2 , and ℓ ∞ metrics. In this paper, we introduce a novel algorithmic framework that blends a general-purpose constrained-optimization solver PyGRANSO, With Constraint-Folding (PWCF), to add reliability and generality to robustness evaluation. PWCF 1) finds good-quality solutions without the need of delicate hyperparameter tuning and 2) can handle more general perturbation types, e.g., modeled by general ℓ p (where p > 0) and perceptual (nonℓ p ) distances, which are inaccessible to existing PGD-based algorithms.