SplitGuard:检测和减轻分割学习中的训练劫持攻击

Ege Erdogan, Alptekin Kupcu, A. E. Cicek
{"title":"SplitGuard:检测和减轻分割学习中的训练劫持攻击","authors":"Ege Erdogan, Alptekin Kupcu, A. E. Cicek","doi":"10.1145/3559613.3563198","DOIUrl":null,"url":null,"abstract":"Distributed deep learning frameworks such as split learning provide great benefits with regards to the computational cost of training deep neural networks and the privacy-aware utilization of the collective data of a group of data-holders. Split learning, in particular, achieves this goal by dividing a neural network between a client and a server so that the client computes the initial set of layers, and the server computes the rest. However, this method introduces a unique attack vector for a malicious server attempting to steal the client's private data: the server can direct the client model towards learning any task of its choice, e.g. towards outputting easily invertible values. With a concrete example already proposed (Pasquini et al., CCS '21), such training-hijacking attacks present a significant risk for the data privacy of split learning clients. In this paper, we propose SplitGuard, a method by which a split learning client can detect whether it is being targeted by a training-hijacking attack or not. We experimentally evaluate our method's effectiveness, compare it with potential alternatives, and discuss in detail various points related to its use. We conclude that SplitGuard can effectively detect training-hijacking attacks while minimizing the amount of information recovered by the adversaries.","PeriodicalId":416548,"journal":{"name":"Proceedings of the 21st Workshop on Privacy in the Electronic Society","volume":"7 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":"{\"title\":\"SplitGuard: Detecting and Mitigating Training-Hijacking Attacks in Split Learning\",\"authors\":\"Ege Erdogan, Alptekin Kupcu, A. E. Cicek\",\"doi\":\"10.1145/3559613.3563198\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Distributed deep learning frameworks such as split learning provide great benefits with regards to the computational cost of training deep neural networks and the privacy-aware utilization of the collective data of a group of data-holders. Split learning, in particular, achieves this goal by dividing a neural network between a client and a server so that the client computes the initial set of layers, and the server computes the rest. However, this method introduces a unique attack vector for a malicious server attempting to steal the client's private data: the server can direct the client model towards learning any task of its choice, e.g. towards outputting easily invertible values. With a concrete example already proposed (Pasquini et al., CCS '21), such training-hijacking attacks present a significant risk for the data privacy of split learning clients. In this paper, we propose SplitGuard, a method by which a split learning client can detect whether it is being targeted by a training-hijacking attack or not. We experimentally evaluate our method's effectiveness, compare it with potential alternatives, and discuss in detail various points related to its use. We conclude that SplitGuard can effectively detect training-hijacking attacks while minimizing the amount of information recovered by the adversaries.\",\"PeriodicalId\":416548,\"journal\":{\"name\":\"Proceedings of the 21st Workshop on Privacy in the Electronic Society\",\"volume\":\"7 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-08-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"14\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 21st Workshop on Privacy in the Electronic Society\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3559613.3563198\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 21st Workshop on Privacy in the Electronic Society","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3559613.3563198","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 14

摘要

分布式深度学习框架,如分裂学习,在训练深度神经网络的计算成本和一组数据持有者的集体数据的隐私意识利用方面提供了巨大的好处。特别是,分裂学习通过在客户端和服务器之间划分神经网络来实现这一目标,以便客户端计算初始层集,服务器计算其余部分。然而,这种方法为试图窃取客户端私有数据的恶意服务器引入了一个独特的攻击向量:服务器可以引导客户端模型学习它选择的任何任务,例如输出容易可逆的值。已经提出了一个具体的例子(Pasquini et al., CCS '21),这种训练劫持攻击对分裂学习客户端的数据隐私构成了重大风险。在本文中,我们提出了SplitGuard,这是一种分裂学习客户端可以检测它是否被训练劫持攻击的方法。我们通过实验评估了我们的方法的有效性,将其与潜在的替代方法进行了比较,并详细讨论了与使用相关的各个要点。我们得出的结论是,SplitGuard可以有效地检测训练劫持攻击,同时最大限度地减少对手恢复的信息量。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
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
SplitGuard: Detecting and Mitigating Training-Hijacking Attacks in Split Learning
Distributed deep learning frameworks such as split learning provide great benefits with regards to the computational cost of training deep neural networks and the privacy-aware utilization of the collective data of a group of data-holders. Split learning, in particular, achieves this goal by dividing a neural network between a client and a server so that the client computes the initial set of layers, and the server computes the rest. However, this method introduces a unique attack vector for a malicious server attempting to steal the client's private data: the server can direct the client model towards learning any task of its choice, e.g. towards outputting easily invertible values. With a concrete example already proposed (Pasquini et al., CCS '21), such training-hijacking attacks present a significant risk for the data privacy of split learning clients. In this paper, we propose SplitGuard, a method by which a split learning client can detect whether it is being targeted by a training-hijacking attack or not. We experimentally evaluate our method's effectiveness, compare it with potential alternatives, and discuss in detail various points related to its use. We conclude that SplitGuard can effectively detect training-hijacking attacks while minimizing the amount of information recovered by the adversaries.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Darwin's Theory of Censorship: Analysing the Evolution of Censored Topics with Dynamic Topic Models PRSONA Tracking the Evolution of Cookie-based Tracking on Facebook Fingerprinting and Personal Information Leakage from Touchscreen Interactions Privacy and Security Evaluation of Mobile Payment Applications Through User-Generated Reviews
×
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