南京邮电大学学报（自然科学版）2026，Vol.46Issue(2)：56-65,10.DOI:10.14132/j.cnki.1673-5439.2026.02.007
基于个性化本地动量的自适应差分隐私联邦学习
Personalized federated learning via adaptive differential privacy with local momentum
摘要
Abstract
Federated learning(FL)is a distributed machine learning paradigm that enables multiple de-vices or organizations to collaboratively train models by sharing training parameters instead of raw data.However,shared parameters remain vulnerable to inference attacks,e.g.differential attacks,which can still compromise individual privacy.To mitigate such threats,differential privacy(DP)has been widely adopted in FL frameworks.This paper investigates a client-level DP federated learning scenario with noise optimization,and proposes an adaptive DP-FL algorithm based on personalized local momentum.Specifically,we first implement dynamic calibration of local models through client-specific momentum mechanisms to address the client drift problem.Furthermore,we design an adaptive DP mechanism fea-turing decaying clipping thresholds,which dynamically optimizes the noise injection scale for individual clients.Experimental evaluations on Cifar10,Cifar100,and SVHN datasets demonstrate that our algo-rithm outperforms existing approaches under equivalent privacy protection levels.关键词
联邦学习/差分隐私/数据异构/动量/自适应噪声Key words
federated learning(FL)/differential privacy(DP)/data heterogeneity/momentum/adap-tive noise分类
信息技术与安全科学引用本文复制引用
杨健,张世召,夏友旭,王藤遇,钱奕安..基于个性化本地动量的自适应差分隐私联邦学习[J].南京邮电大学学报（自然科学版）,2026,46(2):56-65,10.基金项目
国家自然科学基金青年基金(62201285)和江苏省研究生科研与实践创新计划(SJCX24_0317)资助项目 （62201285）
