English

Debunking Optimization Myths in Federated Learning for Medical Image Classification

Machine Learning 2025-07-29 v1 Image and Video Processing Signal Processing

Abstract

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.

Keywords

Cite

@article{arxiv.2507.19822,
  title  = {Debunking Optimization Myths in Federated Learning for Medical Image Classification},
  author = {Youngjoon Lee and Hyukjoon Lee and Jinu Gong and Yang Cao and Joonhyuk Kang},
  journal= {arXiv preprint arXiv:2507.19822},
  year   = {2025}
}

Comments

Accepted to Efficient Medical AI Workshop - MICCAI 2025

R2 v1 2026-07-01T04:19:55.804Z