Saved in:
Bibliographic Details
Main Authors: Sun, Yuxia, Sun, Aoxiang, Pan, Siyi, Fu, Zhixiao, Guo, Jingcai
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.07456
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912233146023936
author Sun, Yuxia
Sun, Aoxiang
Pan, Siyi
Fu, Zhixiao
Guo, Jingcai
author_facet Sun, Yuxia
Sun, Aoxiang
Pan, Siyi
Fu, Zhixiao
Guo, Jingcai
contents Personalized federated learning (PFL) tailors models to clients' unique data distributions while preserving privacy. However, existing aggregation-weight-based PFL methods often struggle with heterogeneous data, facing challenges in accuracy, computational efficiency, and communication overhead. We propose FedAPA, a novel PFL method featuring a server-side, gradient-based adaptive aggregation strategy to generate personalized models, by updating aggregation weights based on gradients of client-parameter changes with respect to the aggregation weights in a centralized manner. FedAPA guarantees theoretical convergence and achieves superior accuracy and computational efficiency compared to 10 PFL competitors across three datasets, with competitive communication overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2502_07456
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FedAPA: Server-side Gradient-Based Adaptive Personalized Aggregation for Federated Learning on Heterogeneous Data
Sun, Yuxia
Sun, Aoxiang
Pan, Siyi
Fu, Zhixiao
Guo, Jingcai
Machine Learning
Computer Vision and Pattern Recognition
Personalized federated learning (PFL) tailors models to clients' unique data distributions while preserving privacy. However, existing aggregation-weight-based PFL methods often struggle with heterogeneous data, facing challenges in accuracy, computational efficiency, and communication overhead. We propose FedAPA, a novel PFL method featuring a server-side, gradient-based adaptive aggregation strategy to generate personalized models, by updating aggregation weights based on gradients of client-parameter changes with respect to the aggregation weights in a centralized manner. FedAPA guarantees theoretical convergence and achieves superior accuracy and computational efficiency compared to 10 PFL competitors across three datasets, with competitive communication overhead.
title FedAPA: Server-side Gradient-Based Adaptive Personalized Aggregation for Federated Learning on Heterogeneous Data
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2502.07456