AdvFlow: Inconspicuous Black-box Adversarial Attacks using Normalizing Flows

TL;DR

AdvFlow is a novel black-box adversarial attack method leveraging normalizing flows to generate imperceptible adversarial examples that closely follow the data distribution, making detection difficult and demonstrating competitive effectiveness.

Contribution

This paper introduces AdvFlow, the first attack method using normalizing flows to produce natural-looking adversarial examples in a black-box setting.

Findings

AdvFlow generates adversaries that mimic the data distribution.

The method achieves competitive attack success rates.

Adversaries are less detectable due to their distributional similarity.

Abstract

Deep learning classifiers are susceptible to well-crafted, imperceptible variations of their inputs, known as adversarial attacks. In this regard, the study of powerful attack models sheds light on the sources of vulnerability in these classifiers, hopefully leading to more robust ones. In this paper, we introduce AdvFlow: a novel black-box adversarial attack method on image classifiers that exploits the power of normalizing flows to model the density of adversarial examples around a given target image. We see that the proposed method generates adversaries that closely follow the clean data distribution, a property which makes their detection less likely. Also, our experimental results show competitive performance of the proposed approach with some of the existing attack methods on defended classifiers. The code is available at https://github.com/hmdolatabadi/AdvFlow.