Defending against Universal Perturbations with Shared Adversarial Training

Classifiers such as deep neural networks have been shown to be vulnerable against adversarial perturbations on problems with high-dimensional input space. While adversarial training improves the robustness of image classifiers against such adversarial perturbations, it leaves them sensitive to perturbations on a non-negligible fraction of the inputs. In this work, we show that adversarial training is more effective in preventing universal perturbations, where the same perturbation needs to fool a classifier on many inputs. Moreover, we investigate the trade-off between robustness against universal perturbations and performance on unperturbed data and propose an extension of adversarial training that handles this trade-off more gracefully. We present results for image classification and semantic segmentation to showcase that universal perturbations that fool a model hardened with adversarial training become clearly perceptible and show patterns of the target scene.

READ FULL TEXT

page 1

page 12

page 13

page 14

page 15

page 16

page 17

page 18

research
10/10/2022

Universal Adversarial Perturbations: Efficiency on a small image dataset

Although neural networks perform very well on the image classification t...
research
01/12/2018

A3T: Adversarially Augmented Adversarial Training

Recent research showed that deep neural networks are highly sensitive to...
research
01/27/2021

Meta Adversarial Training

Recently demonstrated physical-world adversarial attacks have exposed vu...
research
09/20/2018

Playing the Game of Universal Adversarial Perturbations

We study the problem of learning classifiers robust to universal adversa...
research
04/28/2023

On the existence of solutions to adversarial training in multiclass classification

We study three models of the problem of adversarial training in multicla...
research
02/21/2020

Robustness from Simple Classifiers

Despite the vast success of Deep Neural Networks in numerous application...
research
05/08/2020

Towards Robustness against Unsuspicious Adversarial Examples

Despite the remarkable success of deep neural networks, significant conc...

Please sign up or login with your details

Forgot password? Click here to reset