Adversarial Examples Detection beyond Image Space

02/23/2021
by   Kejiang Chen, et al.
0

Deep neural networks have been proved that they are vulnerable to adversarial examples, which are generated by adding human-imperceptible perturbations to images. To defend these adversarial examples, various detection based methods have been proposed. However, most of them perform poorly on detecting adversarial examples with extremely slight perturbations. By exploring these adversarial examples, we find that there exists compliance between perturbations and prediction confidence, which guides us to detect few-perturbation attacks from the aspect of prediction confidence. To detect both few-perturbation attacks and large-perturbation attacks, we propose a method beyond image space by a two-stream architecture, in which the image stream focuses on the pixel artifacts and the gradient stream copes with the confidence artifacts. The experimental results show that the proposed method outperforms the existing methods under oblivious attacks and is verified effective to defend omniscient attacks as well.

READ FULL TEXT

page 1

page 3

research
05/06/2023

Reactive Perturbation Defocusing for Textual Adversarial Defense

Recent studies have shown that large pre-trained language models are vul...
research
06/08/2019

ML-LOO: Detecting Adversarial Examples with Feature Attribution

Deep neural networks obtain state-of-the-art performance on a series of ...
research
10/27/2019

Spot Evasion Attacks: Adversarial Examples for License Plate Recognition Systems with Convolution Neural Networks

Recent studies have shown convolution neural networks (CNNs) for image r...
research
12/15/2018

Perturbation Analysis of Learning Algorithms: A Unifying Perspective on Generation of Adversarial Examples

Despite the tremendous success of deep neural networks in various learni...
research
02/22/2023

Mitigating Adversarial Attacks in Deepfake Detection: An Exploration of Perturbation and AI Techniques

Deep learning is a crucial aspect of machine learning, but it also makes...
research
04/03/2019

Interpreting Adversarial Examples by Activation Promotion and Suppression

It is widely known that convolutional neural networks (CNNs) are vulnera...
research
07/31/2020

TEAM: We Need More Powerful Adversarial Examples for DNNs

Although deep neural networks (DNNs) have achieved success in many appli...

Please sign up or login with your details

Forgot password? Click here to reset