Omni-Detection of Adversarial Examples with Diverse Magnitudes  

作　　者：Ke Jianpeng Wang Wenqi Yang Kang Wang Lina Ye Aoshuang Wang Run 

机构地区：[1]Key Laboratory of Aerospace Information Security and Trusted Computing(Wuhan University),Ministry of Education,Wuhan 430072,China [2]School of Cyber Science and Engineering,Wuhan University,Wuhan 430072,China

出　　处：《China Communications》2024年第12期139-151,共13页中国通信（英文版）

基　　金：This work was partly supported by the National Natural Science Foundation of China under No.62372334,61876134,and U1836112.

摘　　要：Deep neural networks(DNNs)are poten-tially susceptible to adversarial examples that are ma-liciously manipulated by adding imperceptible pertur-bations to legitimate inputs,leading to abnormal be-havior of models.Plenty of methods have been pro-posed to defend against adversarial examples.How-ever,the majority of them are suffering the follow-ing weaknesses:1)lack of generalization and prac-ticality.2)fail to deal with unknown attacks.To ad-dress the above issues,we design the adversarial na-ture eraser(ANE)and feature map detector(FMD)to detect fragile and high-intensity adversarial examples,respectively.Then,we apply the ensemble learning method to compose our detector,dealing with adver-sarial examples with diverse magnitudes in a divide-and-conquer manner.Experimental results show that our approach achieves 99.30%and 99.62%Area un-der Curve(AUC)scores on average when tested with various Lp norm-based attacks on CIFAR-10 and Im-ageNet,respectively.Furthermore,our approach also shows its potential in detecting unknown attacks.

关 键 词：adversarial example detection ensemble learning feature maps fragile and high-intensity ad-versarial examples 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]