mFI-PSO: A Flexible and Effective Method in Adversarial Image Generation for Deep Neural Networks

TL;DR

The paper introduces mFI-PSO, a novel method combining manifold-based influence measures and particle swarm optimization to generate effective and customizable adversarial images against deep neural networks.

Contribution

It presents a new approach that improves adversarial image generation by integrating influence measures with optimization, offering greater flexibility and effectiveness.

Findings

mFI-PSO outperforms existing methods in attack success rate

It allows customizable adversarial perturbations

Demonstrates high effectiveness across different DNN models

Abstract

Deep neural networks (DNNs) have achieved great success in image classification, but can be very vulnerable to adversarial attacks with small perturbations to images. To improve adversarial image generation for DNNs, we develop a novel method, called mFI-PSO, which utilizes a Manifold-based First-order Influence measure for vulnerable image and pixel selection and the Particle Swarm Optimization for various objective functions. Our mFI-PSO can thus effectively design adversarial images with flexible, customized options on the number of perturbed pixels, the misclassification probability, and the targeted incorrect class. Experiments demonstrate the flexibility and effectiveness of our mFI-PSO in adversarial attacks and its appealing advantages over some popular methods.