Universal Adversarial Perturbations: Efficiency on a small image dataset

TL;DR

This paper investigates the effectiveness of universal adversarial perturbations on small neural networks and datasets, aiming to understand their efficiency and potential vulnerabilities in constrained settings.

Contribution

It reproduces the universal adversarial perturbations experiment on smaller models and datasets to analyze their efficiency and robustness.

Findings

Universal perturbations can fool small neural networks.

Efficiency of perturbations varies with network size and dataset complexity.

Provides insights into adversarial vulnerability in resource-constrained scenarios.

Abstract

Although neural networks perform very well on the image classification task, they are still vulnerable to adversarial perturbations that can fool a neural network without visibly changing an input image. A paper has shown the existence of Universal Adversarial Perturbations which when added to any image will fool the neural network with a very high probability. In this paper we will try to reproduce the experience of the Universal Adversarial Perturbations paper, but on a smaller neural network architecture and training set, in order to be able to study the efficiency of the computed perturbation.