A comprehensive survey of robust deep learning in computer vision

作     者：Jia Liu Yaochu Jin 

作者机构：Ping An Property&Casualty Insurance CompanyShenzhen518048GuangdongChina School of EngineeringWestlake UniversityHangzhou310030China 

出 版 物：《Journal of Automation and Intelligence》 (自动化与人工智能（英文）)

年 卷 期：2023年第2卷第4期

页      面：175-195页

学科分类：0502[文学-外国语言文学] 050201[文学-英语语言文学] 05[文学] 

基　　金：Alexander von Humboldt-Stiftung, AvH Bundesministerium für Bildung und Forschung, BMBF 

主　　题：Robustness Deep learning Computer vision Survey Adversarial attack Adversarial defenses 

摘      要：Deep learning has presented remarkable progress in various *** the excellent performance,deep learning models remain not robust,especially to well-designed adversarial examples,limiting deep learning models employed in security-critical ***,how to improve the robustness of deep learning has attracted increasing attention from *** paper investigates the progress on the threat of deep learning and the techniques that can enhance the model robustness in computer *** previous relevant survey papers summarizing adversarial attacks and defense technologies,this paper also provides an overview of the general robustness of deep ***,this survey elaborates on the current robustness evaluation approaches,which require further *** paper also reviews the recent literature on making deep learning models resistant to adversarial examples from an architectural perspective,which was rarely mentioned in previous ***,interesting directions for future research are listed based on the reviewed *** survey is hoped to serve as the basis for future research in this topical field.