An Alternative Surrogate Loss for PGD-based Adversarial Testing

Abstract

Adversarial testing methods based on Projected Gradient Descent (PGD) are widely used for searching norm-bounded perturbations that cause the inputs of neural networks to be misclassified. This paper takes a deeper look at these methods and explains the effect of different hyperparameters (i.e., optimizer, step size and surrogate loss). We introduce the concept of MultiTargeted testing, which makes clever use of alternative surrogate losses, and explain when and how MultiTargeted is guaranteed to find optimal perturbations. Finally, we demonstrate that MultiTargeted outperforms more sophisticated methods and often requires less iterative steps than other variants of PGD found in the literature. Notably, MultiTargeted ranks first on MadryLab's white-box MNIST and CIFAR-10 leaderboards, reducing the accuracy of their MNIST model to 88.36% (with ${\ell }_{\infty }$ perturbations of $\epsilon =0.3$) and the accuracy of their CIFAR-10 model to 44.03% (at $\epsilon =8/255$). MultiTargeted also ranks first on the TRADES leaderboard reducing the accuracy of their CIFAR-10 model to 53.07% (with ${\ell }_{\infty }$ perturbations of $\epsilon =0.031$).

References in corpus (4)

• Explaining and Harnessing Adversarial Examples
• Theoretically Principled Trade-off between Robustness and Accuracy
• On Evaluating Adversarial Robustness
• Enhancing Gradient-based Attacks with Symbolic Intervals

Cited by in corpus (12)

• Adversarial Examples Make Strong Poisons
• Attacking Adversarial Attacks as A Defense
• Label Smoothing and Adversarial Robustness
• Adversarial Learning Guarantees for Linear Hypotheses and Neural Networks
• Robust and Private Learning of Halfspaces
• LAFEAT: Piercing Through Adversarial Defenses with Latent Features
• Adversarial Attacks on ML Defense Models Competition
• Model-Agnostic Meta-Attack: Towards Reliable Evaluation of Adversarial Robustness
• Improving Hierarchical Adversarial Robustness of Deep Neural Networks
• A Multiclass Boosting Framework for Achieving Fast and Provable Adversarial Robustness
• Towards Sharper First-Order Adversary with Quantized Gradients
• A Closer Look at the Adversarial Robustness of Information Bottleneck Models