联邦学习隐私模型发布综述
DOI:
作者:
作者单位:

1.信息网络安全国网重点实验室 全球能源互联网研究院有限公司南京分公司;2.南京大学

作者简介:

通讯作者:

中图分类号:

基金项目:

国家电网有限公司总部管理科技项目资助:基于联邦学习的电力数据安全协同技术研究(项目编码:5700-202190184A-0-0-00)


Survey on Private Model Publishing of Federated Learning
Author:
Affiliation:

1.State Grid Key Laboratory of Information &2.Network Security, Global Energy Interconnection Research Institute Nanjing Branch;3.Nanjing University

Fund Project:

This work is supported by the science and technology project of State Grid Corporation of China:”Research on Power Data Security Collaboration Technology Based on Federated Learning”(Grand No. 5700-202190184A-0-0-00)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    联邦学习这一类分布式机器学习技术旨在保证使用大数据进行机器学习训练时保护本地数据不泄露。然而一系列机器学习隐私攻击表明,即使不直接暴露本地数据,仅仅通过获取机器学习模型的参数就可以进行数据隐私的窃取。从训练时参与者和聚合端之间传递的中间模型到最后发布的聚合模型,联邦学习的模型发布过程存在着诸多的隐私威胁。由此也出现了大量相关的保护技术,包括基于差分隐私以及基于密码学的联邦学习隐私保护技术。本文针对联邦学习本地模型和聚合模型发布过程中可能出现的各种隐私威胁和敌手模型进行了简要介绍,并且对相关的防御技术和研究成果进行系统性综述。同时也对相关技术在联邦学习隐私保护中的发展趋势进行了展望。

    Abstract:

    Federated learning is a kind of distributed machine learning technology that ensures that local data is not compromised when training with big data for machine learning models. However, a series of attacks shows that the adversary can steal private information from machine learning model parameters even if local data is inaccessible. Thus, from the intermediate model parameters transmission between the participants and the aggregator in the training phase to the finally released aggregated model, there are many privacy threats during the model release process of federated learning. As a result, many privacy-preserving federated learning approaches have emerged, primarily based on cryptography and differential privacy technology. This paper surveys the various privacy threats and adversary models that may appear when we publish local models and the aggregated model of federated learning. Furthermore, we systematically summarize the related defense technologies and research results. Additionally, we also look forward to the development of privacy-preserving federated learning.

    参考文献
    相似文献
    引证文献
引用本文

石聪聪,高先周,黄秀丽,毛云龙.联邦学习隐私模型发布综述[J].南京信息工程大学学报,,():

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-10-20
  • 最后修改日期:2021-10-28
  • 录用日期:2021-10-28
  • 在线发布日期:
  • 出版日期:

地址:江苏南京,宁六路219号,南京信息工程大学    邮编:210044

联系电话:025-58731025    E-mail:nxdxb@nuist.edu.cn

南京信息工程大学学报 ® 2022 版权所有  技术支持:北京勤云科技发展有限公司