Identification of Attack-Specific Signatures in Adversarial Examples
This work addresses the need for more nuanced comparisons of adversarial attacks beyond success rates, which is incremental for researchers in adversarial machine learning.
The paper tackled the problem of distinguishing between different adversarial attack algorithms by showing that each produces distinct signatures in adversarial examples, both in terms of the attack's identity and its qualitative impact on neural networks. They demonstrated this by identifying the attack algorithm from examples and using saliency maps to reveal differences in targeted network parts and image regions.
The adversarial attack literature contains a myriad of algorithms for crafting perturbations which yield pathological behavior in neural networks. In many cases, multiple algorithms target the same tasks and even enforce the same constraints. In this work, we show that different attack algorithms produce adversarial examples which are distinct not only in their effectiveness but also in how they qualitatively affect their victims. We begin by demonstrating that one can determine the attack algorithm that crafted an adversarial example. Then, we leverage recent advances in parameter-space saliency maps to show, both visually and quantitatively, that adversarial attack algorithms differ in which parts of the network and image they target. Our findings suggest that prospective adversarial attacks should be compared not only via their success rates at fooling models but also via deeper downstream effects they have on victims.