A Closer Look at Personalized Fine-Tuning in Heterogeneous Federated Learning

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Chen, Minghui, Ghoukasian, Hrad, Jin, Ruinan, Wang, Zehua, Karimireddy, Sai Praneeth, Li, Xiaoxiao
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909977333989376
author Chen, Minghui
Ghoukasian, Hrad
Jin, Ruinan
Wang, Zehua
Karimireddy, Sai Praneeth
Li, Xiaoxiao
author_facet Chen, Minghui
Ghoukasian, Hrad
Jin, Ruinan
Wang, Zehua
Karimireddy, Sai Praneeth
Li, Xiaoxiao
contents Federated Learning (FL) enables decentralized, privacy-preserving model training but struggles to balance global generalization and local personalization due to non-identical data distributions across clients. Personalized Fine-Tuning (PFT), a popular post-hoc solution, fine-tunes the final global model locally but often overfits to skewed client distributions or fails under domain shifts. We propose adapting Linear Probing followed by full Fine-Tuning (LP-FT), a principled centralized strategy for alleviating feature distortion (Kumar et al., 2022), to the FL setting. Through systematic evaluation across seven datasets and six PFT variants, we demonstrate LP-FT's superiority in balancing personalization and generalization. Our analysis uncovers federated feature distortion, a phenomenon where local fine-tuning destabilizes globally learned features, and theoretically characterizes how LP-FT mitigates this via phased parameter updates. We further establish conditions (e.g., partial feature overlap, covariate-concept shift) under which LP-FT outperforms standard fine-tuning, offering actionable guidelines for deploying robust personalization in FL.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12695
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Closer Look at Personalized Fine-Tuning in Heterogeneous Federated Learning
Chen, Minghui
Ghoukasian, Hrad
Jin, Ruinan
Wang, Zehua
Karimireddy, Sai Praneeth
Li, Xiaoxiao
Machine Learning
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
Federated Learning (FL) enables decentralized, privacy-preserving model training but struggles to balance global generalization and local personalization due to non-identical data distributions across clients. Personalized Fine-Tuning (PFT), a popular post-hoc solution, fine-tunes the final global model locally but often overfits to skewed client distributions or fails under domain shifts. We propose adapting Linear Probing followed by full Fine-Tuning (LP-FT), a principled centralized strategy for alleviating feature distortion (Kumar et al., 2022), to the FL setting. Through systematic evaluation across seven datasets and six PFT variants, we demonstrate LP-FT's superiority in balancing personalization and generalization. Our analysis uncovers federated feature distortion, a phenomenon where local fine-tuning destabilizes globally learned features, and theoretically characterizes how LP-FT mitigates this via phased parameter updates. We further establish conditions (e.g., partial feature overlap, covariate-concept shift) under which LP-FT outperforms standard fine-tuning, offering actionable guidelines for deploying robust personalization in FL.
title A Closer Look at Personalized Fine-Tuning in Heterogeneous Federated Learning
topic Machine Learning
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2511.12695