Generation Method of Error-Specific Adversarial Examples Using Gradient Information for the Target Class

Ryo Kumagai, S. Takemoto, Y. Nozaki, M. Yoshikawa
{"title":"Generation Method of Error-Specific Adversarial Examples Using Gradient Information for the Target Class","authors":"Ryo Kumagai, S. Takemoto, Y. Nozaki, M. Yoshikawa","doi":"10.1109/ICIET56899.2023.10111283","DOIUrl":null,"url":null,"abstract":"With the advancement of AI technology, vulnerabilities of AI systems have been pointed out. Adversarial Examples (AEs), in which makes AI wrong decisions, are one of the dreaded attacks for AI. Therefore, a thorough investigation of AEs is essential for the safe use of AI. In this paper, we propose a method for generating adversarial examples using gradient information for the target class of input images. We experimentally prove that the proposed method can generate target AEs that misclassify to an arbitrary class with high probability.","PeriodicalId":332586,"journal":{"name":"2023 11th International Conference on Information and Education Technology (ICIET)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 11th International Conference on Information and Education Technology (ICIET)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICIET56899.2023.10111283","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

With the advancement of AI technology, vulnerabilities of AI systems have been pointed out. Adversarial Examples (AEs), in which makes AI wrong decisions, are one of the dreaded attacks for AI. Therefore, a thorough investigation of AEs is essential for the safe use of AI. In this paper, we propose a method for generating adversarial examples using gradient information for the target class of input images. We experimentally prove that the proposed method can generate target AEs that misclassify to an arbitrary class with high probability.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
基于梯度信息的目标类特定误差对抗示例生成方法
随着人工智能技术的进步,人工智能系统的漏洞被指出。对抗性示例(ae),即让AI做出错误决策,是AI最可怕的攻击之一。因此,对人工智能的安全使用进行彻底的调查是必不可少的。在本文中,我们提出了一种使用梯度信息为输入图像的目标类生成对抗示例的方法。实验证明,该方法可以高概率地生成错误分类到任意类别的目标ae。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Using Deep Learning to Track Representational Flexibility Development of Children with Autism in a Virtual World Emotional Responses toward the Flipped Classroom Approach across Academic Disciplines Assessment of Learning Outcomes in the (Java) Object-Oriented Programming Courses Human’s Reaction Time Based Score Calculation of Self-practice Dynamic Yoga System for User’s Feedback by OpenPose and Fuzzy Rules Using 360 Virtual Reality Video in History Learning
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1