ScholarMate
English简体中文繁體中文
客服热线:400-1616-289
登录注册
期刊论文

A Differentially Private Federated Learning Model Against Poisoning Attacks in Edge Computing

Zhou, Jun; Wu, Nan; Wang, Yisong; Gu, Shouzhen*; Cao, Zhenfu; Dong, Xiaolei; Choo, Kim-Kwang Raymond
IEEE Transactions on Dependable and Secure Computing2023-5DOI: 10.1109/TDSC.2022.3168556
10被引7阅读
引用收藏
分享

摘要

Federated learning is increasingly popular, as it allows us to circumvent challenges due to data islands (e.g., challenges in getting access to raw datasets, particularly when data are owned by multiple data owners) by training a global model using data from one or more data owners / sources. However, in an edge computing deployment one cannot assume that resource-constrained end devices are sufficiently secure. In other words, we have to consider the possibility of these devices being compromised and can be abused to facilitate poisoning attacks. Privacy-preserving is another important property to consider when dealing with sensitive user data on end devices. Most existing approaches only consider either defending against poisoning attacks or supporting privacy, but not both properties simultaneously. In this paper, we propose a differentially private federated learning model against poisoning attacks, designed for edge computing deployment. First, we design a weight-based algorithm to perform anomaly detection on the parameters uploaded by end devices in edge nodes, which improves detection rate using only small-size validation datasets and minimizes the communication cost. Then, differential privacy technology is leveraged to protect the privacy of both data and model in an edge computing setting. We also evaluate the detection performance in the presence of random and customized malicious end devices, and compare its performance with those of two other competing approaches published in IEEE Transactions on Dependable and Secure Computing, in terms of attack resiliency, and communication and computation costs. Experimental results demonstrate that our scheme can achieve an optimal tradeoff between security, efficiency and accuracy.

学科与关键词

关键词
Computational modelingPrivacyEdge computingTrainingData modelsCollaborative workImage edge detectionDifferential privacyedge computingfederated learninghigh-practicabilitypoisoning attack
相关
加载中…
  • APFed: Anti-Poisoning Attacks in Privacy-Preserving Heterogeneous Federated Learning
    Chen, Xiao; Yu, Haining*; Jia, Xiaohua; Yu, XiangzhanIEEE Transactions on Information Forensics and Security, 2023, 18: 5749-5761.被引 1
  • Privacy Preservation for Federated Learning With Robust Aggregation in Edge Computing
    Liu, Wentao; Xu, Xiaolong*; Li, Dejuan; Qi, Lianyong; Dai, Fei; Dou, Wanchun; Ni, QiangIEEE Internet of Things Journal, 2023, 10(8): 7343-7355.被引 14
  • Communication-Efficient and Attack-Resistant Federated Edge Learning With Dataset Distillation
    Zhou, Yanlin; Ma, Xiyao; Wu, Dapeng*; Li, Xiaolin*IEEE Transactions on Cloud Computing, 2023, 11(3): 2517-2528.
查看全部 →

发表信息

论文状态
公开发表
期刊
IEEE Transactions on Dependable and Secure Computing
发表日期
2023-05
卷期
第 20 卷, 第 3 期
页码
1941-1958
DOI
10.1109/TDSC.2022.3168556
语言
ENGLISH
成果收录
SCIE
资助标注
National Natural Science Foundation of China [62172161, 62132005, 62172162]; Natural Science Foundation of Shanghai [20ZR1418400]; Cloud Technology Endowed Professorship

作者

  • Z

    Zhou, Jun (第一作者)

  • W

    Wu, Nan

查看全部 7 位作者收起作者
  • W

    Wang, Yisong

  • G

    Gu, Shouzhen (通讯)

  • C

    Cao, Zhenfu

  • D

    Dong, Xiaolei

  • C

    Choo, Kim-Kwang Raymond

登录解锁更多功能

  • 收藏这篇,随时在“我的 · 收藏”找回
  • 认领你的成果,生成推广素材
  • 关注作者获取最新成果
  • 加入项目组、协同写作
注册 / 登录

产品服务

  • 科研之友
  • 创新城
  • 科创云

服务支持

  • 帮助中心
  • 隐私政策
  • 服务条款

联系方式

在线客服:【】
客服热线:400-1616-289
电子邮箱:support@scholarmate.com

关注或下载科研之友

微信二维码
微信公众号
客户端下载二维码
下载客户端
科研成果科研人员科研机构科研动态爱瑞思软件

©2026 深圳市科研之友网络服务有限公司

公安备案图标粤公网安备 44030502000213
粤ICP备 16046710 号粤B2-20110417