Federated Learning and Class Imbalances

TL;DR

This paper evaluates the robustness of the RHFL+ federated learning method against class imbalances, extending its application to medical imaging datasets and providing a scalable, deployment-ready implementation.

Contribution

It reproduces and benchmarks RHFL+ with other algorithms, extends its application to real-world medical datasets, and develops a scalable implementation using NVFlare.

Findings

RHFL+ shows robustness under class imbalances

Extended to medical imaging datasets with positive results

Scalable implementation demonstrated across many clients

Abstract

Federated Learning (FL) enables collaborative model training across decentralized devices while preserving data privacy. However, real-world FL deployments face critical challenges such as data imbalances, including label noise and non-IID distributions. RHFL+, a state-of-the-art method, was proposed to address these challenges in settings with heterogeneous client models. This work investigates the robustness of RHFL+ under class imbalances through three key contributions: (1) reproduction of RHFL+ along with all benchmark algorithms under a unified evaluation framework; (2) extension of RHFL+ to real-world medical imaging datasets, including CBIS-DDSM, BreastMNIST and BHI; (3) a novel implementation using NVFlare, NVIDIA's production-level federated learning framework, enabling a modular, scalable and deployment-ready codebase. To validate effectiveness, extensive ablation studies, algorithmic comparisons under various noise conditions and scalability experiments across increasing numbers of clients are conducted.