Feature Squeezing Mitigates and Detects Carlini/Wagner Adversarial Examples

TL;DR

Feature squeezing enhances the robustness of deep learning models against Carlini/Wagner adversarial attacks and aids in their detection, offering a simple yet effective defense mechanism.

Contribution

This paper demonstrates that feature squeezing techniques are effective against the powerful Carlini/Wagner adversarial attacks, extending previous work to new attack methods.

Findings

Feature squeezing significantly reduces success rate of Carlini/Wagner attacks.

It improves detection of adversarial examples.

Enhances model robustness with simple preprocessing.

Abstract

Feature squeezing is a recently-introduced framework for mitigating and detecting adversarial examples. In previous work, we showed that it is effective against several earlier methods for generating adversarial examples. In this short note, we report on recent results showing that simple feature squeezing techniques also make deep learning models significantly more robust against the Carlini/Wagner attacks, which are the best known adversarial methods discovered to date.