Adap DP-FL: Differentially Private Federated Learning with Adaptive Noise

  • Jie Fu
  • , Zhili Chen*
  • , Xiao Han
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

53 Scopus citations

Abstract

Federated learning seeks to address the issue of isolated data islands by making clients disclose only their local training models. However, it was demonstrated that private information could still be inferred by analyzing local model parameters, such as deep neural network model weights. Recently, differential privacy has been applied to federated learning to protect data privacy, but the noise added may degrade the learning performance much. Typically, in previous work, training parameters were clipped equally and noises were added uniformly. The heterogeneity and convergence of training parameters were simply not considered. In this paper, we propose a differentially private scheme for federated learning with adaptive noise (Adap DP-FL). Specifically, due to the gradient heterogeneity, we conduct adaptive gradient clipping for different clients and different rounds; due to the gradient convergence, we add decreasing noises accordingly. Extensive experiments on real-world datasets demonstrate that our Adap DP-FL outperforms previous methods significantly.

Original languageEnglish
Title of host publicationProceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages656-663
Number of pages8
ISBN (Electronic)9781665494250
DOIs
StatePublished - 2022
Event21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022 - Virtual, Online, China
Duration: 9 Dec 202211 Dec 2022

Publication series

NameProceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022

Conference

Conference21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022
Country/TerritoryChina
CityVirtual, Online
Period9/12/2211/12/22

Keywords

  • adaptive noise
  • differential privacy
  • edge compute
  • federated learning
  • information security and privacy

Fingerprint

Dive into the research topics of 'Adap DP-FL: Differentially Private Federated Learning with Adaptive Noise'. Together they form a unique fingerprint.

Cite this