Feature Squeezing Mitigates and Detects Carlini/Wagner Adversarial Examples

By Weilin Xu, David E. Evans, Yanjun Qi. 2017

47 citations recorded by the source corpus.

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.

Subfields: Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Safety, Robustness & Interpretability

Citations in the corpus

ZOO: Zeroth Order Optimization Based Black-box Attacks to Deep Neural Networks without Training Substitute Models (2017, background): “readers can refer to recent works in [10, 13, 18, 28, 42, 43] and references therein for details.”
Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks (2017, methodology): “paper [42, 43]; we shared details of our approach and code with the authors”
Adversarial Example Defenses: Ensembles of Weak Defenses are not Strong (2017, methodology): “50% on CIFAR-10 against a static adversary using a variety of Carlini and Wagner’s attacks [31].”
Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey (2018, background): “Whereas [43] demonstrated the effectiveness of this approach against more classical attacks [23], a follow-up work [139] also claims that themethodworks reasonablywell against the more powerful C&W attacks [36].”
Adversarial Attacks and Defenses in Deep Learning (2020, methodology): “[76] further show that the feature-squeezing methods proposed in Ref.”