Robust-PFedproto: robust federated prototype learning based on personalized layers

Federated learning (FL), a distributed machine learning framework, was recognized for retaining training data on remote clients. However, two critical challenges were identified. First, heterogeneous data distributions were commonly observed across clients, which significantly degraded overall train...

Full description

Saved in:
Bibliographic Details
Main Authors: XU Mingdi, LI Zhengxiao, WANG Zihang, JIN Chaoyang
Format: Article
Language:English
Published: POSTS&TELECOM PRESS Co., LTD 2025-06-01
Series:网络与信息安全学报
Subjects:
Online Access:http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2025032
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849709295737765888
author XU Mingdi
LI Zhengxiao
WANG Zihang
JIN Chaoyang
author_facet XU Mingdi
LI Zhengxiao
WANG Zihang
JIN Chaoyang
author_sort XU Mingdi
collection DOAJ
description Federated learning (FL), a distributed machine learning framework, was recognized for retaining training data on remote clients. However, two critical challenges were identified. First, heterogeneous data distributions were commonly observed across clients, which significantly degraded overall training efficiency. Second, the central server could’t access authentic remote client data, allowing adversarial clients to upload malicious model updates, thus negatively affecting global training accuracy and efficiency. To address data heterogeneity, the PFedproto framework was proposed, incorporating personalized layers based on prototype learning. These layers, positioned before each client’s decision layer, were optimized to enhance client model adaptation to local datasets and improve localized task prediction accuracy. Building on this framework, a three-stage defense scheme named RobustPFedproto was developed to strengthen the PFedproto framework’s robustness against data poisoning attacks. Experiments were conducted on four real-world image classification datasets. Results demonstrate that the PFedproto framework effectively mitigates data heterogeneity impacts. Additionally, the Robust-PFedproto scheme not only efficiently handles data heterogeneity but also shows strong robustness against data poisoning attacks.
format Article
id doaj-art-2b88a5ce34b840d4a53c493ba02e5a8b
institution DOAJ
issn 2096-109X
language English
publishDate 2025-06-01
publisher POSTS&TELECOM PRESS Co., LTD
record_format Article
series 网络与信息安全学报
spelling doaj-art-2b88a5ce34b840d4a53c493ba02e5a8b2025-08-20T03:15:20ZengPOSTS&TELECOM PRESS Co., LTD网络与信息安全学报2096-109X2025-06-01116780113011843Robust-PFedproto: robust federated prototype learning based on personalized layersXU MingdiLI ZhengxiaoWANG ZihangJIN ChaoyangFederated learning (FL), a distributed machine learning framework, was recognized for retaining training data on remote clients. However, two critical challenges were identified. First, heterogeneous data distributions were commonly observed across clients, which significantly degraded overall training efficiency. Second, the central server could’t access authentic remote client data, allowing adversarial clients to upload malicious model updates, thus negatively affecting global training accuracy and efficiency. To address data heterogeneity, the PFedproto framework was proposed, incorporating personalized layers based on prototype learning. These layers, positioned before each client’s decision layer, were optimized to enhance client model adaptation to local datasets and improve localized task prediction accuracy. Building on this framework, a three-stage defense scheme named RobustPFedproto was developed to strengthen the PFedproto framework’s robustness against data poisoning attacks. Experiments were conducted on four real-world image classification datasets. Results demonstrate that the PFedproto framework effectively mitigates data heterogeneity impacts. Additionally, the Robust-PFedproto scheme not only efficiently handles data heterogeneity but also shows strong robustness against data poisoning attacks.http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2025032federated learningprototype learningkernel density estimationpoisoning attack
spellingShingle XU Mingdi
LI Zhengxiao
WANG Zihang
JIN Chaoyang
Robust-PFedproto: robust federated prototype learning based on personalized layers
网络与信息安全学报
federated learning
prototype learning
kernel density estimation
poisoning attack
title Robust-PFedproto: robust federated prototype learning based on personalized layers
title_full Robust-PFedproto: robust federated prototype learning based on personalized layers
title_fullStr Robust-PFedproto: robust federated prototype learning based on personalized layers
title_full_unstemmed Robust-PFedproto: robust federated prototype learning based on personalized layers
title_short Robust-PFedproto: robust federated prototype learning based on personalized layers
title_sort robust pfedproto robust federated prototype learning based on personalized layers
topic federated learning
prototype learning
kernel density estimation
poisoning attack
url http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2025032
work_keys_str_mv AT xumingdi robustpfedprotorobustfederatedprototypelearningbasedonpersonalizedlayers
AT lizhengxiao robustpfedprotorobustfederatedprototypelearningbasedonpersonalizedlayers
AT wangzihang robustpfedprotorobustfederatedprototypelearningbasedonpersonalizedlayers
AT jinchaoyang robustpfedprotorobustfederatedprototypelearningbasedonpersonalizedlayers