Aug-ILA: More Transferable Intermediate Level Attacks with Augmented References

Chiu Wai Yan, Dit-Yan Yeung

Keywords: adversarial examples, adversarial transferability, intermediate feature, image augmentation

Abstract: An intriguing property of deep neural networks is that adversarial attacks can transfer across different models. Existing methods such as the Intermediate Level Attack (ILA) further improve black-box transferability by fine-tuning a reference adversarial attack, so as to maximize the perturbation on a pre-specified layer of the source model. In this paper, we revisit ILA and evaluate the effect of applying augmentation to the images before passing them to ILA. We start by looking into the effect of common image augmentation techniques and exploring novel augmentation with the aid of adversarial perturbations. Based on the observations, we propose Aug-ILA, an improved method that enhances the transferability of an existing attack under the ILA framework. Specifically, Aug-ILA has three main characteristics: typical image augmentation such as random cropping and resizing applied to all ILA inputs, reverse adversarial update on the clean image, and interpolation between two attacks on the reference image. Our experimental results show that Aug-ILA outperforms ILA and its subsequent variants, as well as state-of-the-art transfer-based attacks, by achieving $96.99\%$ and $87.84\%$ average attack success rates with perturbation budgets $0.05$ and $0.03$, respectively, on nine undefended models.

One-sentence Summary: We present a fine-tuning method that largely enhances the black-box transferability of adversarial examples.