EnsembleFool: A method to generate adversarial examples based on model fusion strategy. Issue 107 (August 2021)

• Record Type: Journal Article
• Authors: Peng, Wenyu; Liu, Renyang; Wang, Ruxin; Cheng, Taining; Wu, Zifeng; Cai, Li; Zhou, Wei
• Is Part Of: Computers & security. Issue 107(2021)
• Publication Date: 2021-08
• Subjects: Deep learning -- Adversarial examples -- Ensemble models -- Self-adaptive -- White-box attack -- Transferability; Computer security -- Periodicals; Electronic data processing departments -- Security measures -- Periodicals; 005.805
• DOI: 10.1016/j.cose.2021.102317
• Languages: English
• ISSNs: 0167-4048

Abstract

Abstract: Deep neural networks have been shown vulnerable to adversarial attacks launched by adversarial examples. These examples' transferability makes an attack in the real-world feasible, which poses a security threat to deep learning. Considering the limited representation capacity of a single deep model, the transferability of an adversarial example generated by a single attack model would cause the failure of attacking other different models. In this paper, we propose a new adversarial attack method, named EnsembleFool, which flexibly integrates multiple models to enhance adversarial examples' transferability. Specifically, the model confidence concerning an input example reveals the risk of a successful attack. In an iterative attacking case, the result of a previous attack could guide us to enforce a new attack that possesses a higher probability of success. Regarding this, we design a series of integration strategies to improve the adversarial examples in each iteration. Extensive experiments on ImageNet indicate that the proposed method has superior attack performance and transferability than state-of-the-art methods.