A Multi-objective Memetic Algorithm for Auto Adversarial Attack Optimization Design

TL;DR

This paper introduces a multi-objective memetic algorithm that automates the design of adversarial attacks to evaluate and improve the robustness of defended deep learning models efficiently, reducing computational costs.

Contribution

It proposes a novel multi-objective memetic algorithm for automatic adversarial attack optimization, incorporating a comprehensive search space and a data selection strategy to lower evaluation costs.

Findings

Effective in generating near-optimal adversarial attacks on defended models.

Reduces computational burden during attack evaluation.

Demonstrates superior performance on CIFAR and ImageNet datasets.

Abstract

The phenomenon of adversarial examples has been revealed in variant scenarios. Recent studies show that well-designed adversarial defense strategies can improve the robustness of deep learning models against adversarial examples. However, with the rapid development of defense technologies, it also tends to be more difficult to evaluate the robustness of the defensed model due to the weak performance of existing manually designed adversarial attacks. To address the challenge, given the defensed model, the efficient adversarial attack with less computational burden and lower robust accuracy is needed to be further exploited. Therefore, we propose a multi-objective memetic algorithm for auto adversarial attack optimization design, which realizes the automatical search for the near-optimal adversarial attack towards defensed models. Firstly, the more general mathematical model of auto adversarial attack optimization design is constructed, where the search space includes not only the attacker operations, magnitude, iteration number, and loss functions but also the connection ways of multiple adversarial attacks. In addition, we develop a multi-objective memetic algorithm combining NSGA-II and local search to solve the optimization problem. Finally, to decrease the evaluation cost during the search, we propose a representative data selection strategy based on the sorting of cross entropy loss values of each images output by models. Experiments on CIFAR10, CIFAR100, and ImageNet datasets show the effectiveness of our proposed method.