Debunking Optimization Myths in Federated Learning for Medical Image Classification

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Lee, Youngjoon, Lee, Hyukjoon, Gong, Jinu, Cao, Yang, Kang, Joonhyuk
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915410964643840
author Lee, Youngjoon
Lee, Hyukjoon
Gong, Jinu
Cao, Yang
Kang, Joonhyuk
author_facet Lee, Youngjoon
Lee, Hyukjoon
Gong, Jinu
Cao, Yang
Kang, Joonhyuk
contents Federated Learning (FL) is a collaborative learning method that enables decentralized model training while preserving data privacy. Despite its promise in medical imaging, recent FL methods are often sensitive to local factors such as optimizers and learning rates, limiting their robustness in practical deployments. In this work, we revisit vanilla FL to clarify the impact of edge device configurations, benchmarking recent FL methods on colorectal pathology and blood cell classification task. We numerically show that the choice of local optimizer and learning rate has a greater effect on performance than the specific FL method. Moreover, we find that increasing local training epochs can either enhance or impair convergence, depending on the FL method. These findings indicate that appropriate edge-specific configuration is more crucial than algorithmic complexity for achieving effective FL.
format Preprint
id arxiv_https___arxiv_org_abs_2507_19822
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Debunking Optimization Myths in Federated Learning for Medical Image Classification
Lee, Youngjoon
Lee, Hyukjoon
Gong, Jinu
Cao, Yang
Kang, Joonhyuk
Machine Learning
Image and Video Processing
Signal Processing
Federated Learning (FL) is a collaborative learning method that enables decentralized model training while preserving data privacy. Despite its promise in medical imaging, recent FL methods are often sensitive to local factors such as optimizers and learning rates, limiting their robustness in practical deployments. In this work, we revisit vanilla FL to clarify the impact of edge device configurations, benchmarking recent FL methods on colorectal pathology and blood cell classification task. We numerically show that the choice of local optimizer and learning rate has a greater effect on performance than the specific FL method. Moreover, we find that increasing local training epochs can either enhance or impair convergence, depending on the FL method. These findings indicate that appropriate edge-specific configuration is more crucial than algorithmic complexity for achieving effective FL.
title Debunking Optimization Myths in Federated Learning for Medical Image Classification
topic Machine Learning
Image and Video Processing
Signal Processing
url https://arxiv.org/abs/2507.19822