FedWCM: Unleashing the Potential of Momentum-based Federated Learning in Long-Tailed Scenarios

Tianle Li, Yongzhi Huang, Linshan Jiang, Qipeng Xie, Chang Liu +3 more
7/20/2025
cs.LG

No official code found. Best implementation: Li-Tian-Le/FedWCM-Supplement (1 star, last commit 2025-07-15).

Abstract

Federated Learning (FL) enables decentralized model training while preserving data privacy. Despite its benefits, FL faces challenges with non-identically distributed (non-IID) data, especially in long-tailed scenarios with imbalanced class samples. Momentum-based FL methods, often used to accelerate FL convergence, struggle with these distributions, resulting in biased models and making FL hard to converge. To understand this challenge, we conduct extensive investigations into this phenomenon, accompanied by a layer-wise analysis of neural network behavior. Based on these insights, we propose FedWCM, a method that dynamically adjusts momentum using global and per-round data to correct directional biases introduced by long-tailed distributions. Extensive experiments show that FedWCM resolves non-convergence issues and outperforms existing methods, enhancing FL's efficiency and effectiveness in handling client heterogeneity and data imbalance.

View on arXivView PDF

Code Implementations(1)

Li-Tian-Le/FedWCM-SupplementCommunity67%

Supplementary Material for "FedWCM: Unleashing the Potential of Momentum-based Federated Learning in Long-Tailed Scenarios"

10Jul 15, 20251 years ago

Similar Papers

Momentum Benefits Non-IID Federated Learning Simply and Provably

86% match2023

FedLF: Adaptive Logit Adjustment and Feature Optimization in Federated Long-Tailed Learning

86% match1 repo42024

Fed-GraB: Federated Long-tailed Learning with Self-Adjusting Gradient Balancer

85% match1 repo162026

Fed-GraB: Federated Long-tailed Learning with Self-Adjusting Gradient Balancer

85% match1 repo162023

FedLoGe: Joint Local and Generic Federated Learning under Long-tailed Data

85% match1 repo192024

Federated Loss Exploration for Improved Convergence on Non-IID Data

85% match1 repo72025

Cite this paper

@article{li2025fedwcm,
  title  = {FedWCM: Unleashing the Potential of Momentum-based Federated Learning in Long-Tailed Scenarios},
  author = {Tianle Li and Yongzhi Huang and Linshan Jiang and Qipeng Xie and Chang Liu and Wenfeng Du and Lu Wang and Kaishun Wu},
  year   = {2025},
  eprint = {2507.14980},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2507.14980v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2507.14980.svg)](https://codeofpaper.com/paper/2507.14980)

Discussion