Evading Adversarial Example Detection Defenses with Orthogonal Projected Gradient Descent

06/28/2021
by   Oliver Bryniarski, et al.
12

Evading adversarial example detection defenses requires finding adversarial examples that must simultaneously (a) be misclassified by the model and (b) be detected as non-adversarial. We find that existing attacks that attempt to satisfy multiple simultaneous constraints often over-optimize against one constraint at the cost of satisfying another. We introduce Orthogonal Projected Gradient Descent, an improved attack technique to generate adversarial examples that avoids this problem by orthogonalizing the gradients when running standard gradient-based attacks. We use our technique to evade four state-of-the-art detection defenses, reducing their accuracy to 0 detection rate.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/23/2018

Stochastic Substitute Training: A Gray-box Approach to Craft Adversarial Examples Against Gradient Obfuscation Defenses

It has been shown that adversaries can craft example inputs to neural ne...
research
06/19/2022

On the Limitations of Stochastic Pre-processing Defenses

Defending against adversarial examples remains an open problem. A common...
research
06/05/2019

Enhancing Gradient-based Attacks with Symbolic Intervals

Recent breakthroughs in defenses against adversarial examples, like adve...
research
09/13/2023

Mitigating Adversarial Attacks in Federated Learning with Trusted Execution Environments

The main premise of federated learning (FL) is that machine learning mod...
research
03/14/2020

Minimum-Norm Adversarial Examples on KNN and KNN-Based Models

We study the robustness against adversarial examples of kNN classifiers ...
research
10/18/2020

FADER: Fast Adversarial Example Rejection

Deep neural networks are vulnerable to adversarial examples, i.e., caref...
research
07/28/2022

A Recommender System for Equitable Public Art Curation and Installation

The placement of art in public spaces can have a significant impact on w...

Please sign up or login with your details

Forgot password? Click here to reset