A Method for Computing Class-wise Universal Adversarial Perturbations

12/01/2019
by   Tejus Gupta, et al.
0

We present an algorithm for computing class-specific universal adversarial perturbations for deep neural networks. Such perturbations can induce misclassification in a large fraction of images of a specific class. Unlike previous methods that use iterative optimization for computing a universal perturbation, the proposed method employs a perturbation that is a linear function of weights of the neural network and hence can be computed much faster. The method does not require any training data and has no hyper-parameters. The attack obtains 34 state-of-the-art deep neural networks on ImageNet and transfers across models. We also study the characteristics of the decision boundaries learned by standard and adversarially trained models to understand the universal adversarial perturbations.

READ FULL TEXT

page 2

page 8

research
10/10/2019

Universal Adversarial Perturbation for Text Classification

Given a state-of-the-art deep neural network text classifier, we show th...
research
07/28/2020

Cassandra: Detecting Trojaned Networks from Adversarial Perturbations

Deep neural networks are being widely deployed for many critical tasks d...
research
12/16/2018

Trust Region Based Adversarial Attack on Neural Networks

Deep Neural Networks are quite vulnerable to adversarial perturbations. ...
research
02/12/2018

Lipschitz-Margin Training: Scalable Certification of Perturbation Invariance for Deep Neural Networks

High sensitivity of neural networks against malicious perturbations on i...
research
11/30/2022

Towards Interpreting Vulnerability of Multi-Instance Learning via Customized and Universal Adversarial Perturbations

Multi-instance learning (MIL) is a great paradigm for dealing with compl...
research
06/12/2022

Consistent Attack: Universal Adversarial Perturbation on Embodied Vision Navigation

Embodied agents in vision navigation coupled with deep neural networks h...
research
09/19/2020

SecDD: Efficient and Secure Method for Remotely Training Neural Networks

We leverage what are typically considered the worst qualities of deep le...

Please sign up or login with your details

Forgot password? Click here to reset