Distributionally Robust Federated Learning with Client Drift Minimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Krouka, Mounssif, Issaid, Chaouki Ben, Bennis, Mehdi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908373366079488
author Krouka, Mounssif
Issaid, Chaouki Ben
Bennis, Mehdi
author_facet Krouka, Mounssif
Issaid, Chaouki Ben
Bennis, Mehdi
contents Federated learning (FL) faces critical challenges, particularly in heterogeneous environments where non-independent and identically distributed data across clients can lead to unfair and inefficient model performance. In this work, we introduce \textit{DRDM}, a novel algorithm that addresses these issues by combining a distributionally robust optimization (DRO) framework with dynamic regularization to mitigate client drift. \textit{DRDM} frames the training as a min-max optimization problem aimed at maximizing performance for the worst-case client, thereby promoting robustness and fairness. This robust objective is optimized through an algorithm leveraging dynamic regularization and efficient local updates, which significantly reduces the required number of communication rounds. Moreover, we provide a theoretical convergence analysis for convex smooth objectives under partial participation. Extensive experiments on three benchmark datasets, covering various model architectures and data heterogeneity levels, demonstrate that \textit{DRDM} significantly improves worst-case test accuracy while requiring fewer communication rounds than existing state-of-the-art baselines. Furthermore, we analyze the impact of signal-to-noise ratio (SNR) and bandwidth on the energy consumption of participating clients, demonstrating that the number of local update steps can be adaptively selected to achieve a target worst-case test accuracy with minimal total energy cost across diverse communication environments.
format Preprint
id arxiv_https___arxiv_org_abs_2505_15371
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Distributionally Robust Federated Learning with Client Drift Minimization
Krouka, Mounssif
Issaid, Chaouki Ben
Bennis, Mehdi
Machine Learning
Federated learning (FL) faces critical challenges, particularly in heterogeneous environments where non-independent and identically distributed data across clients can lead to unfair and inefficient model performance. In this work, we introduce \textit{DRDM}, a novel algorithm that addresses these issues by combining a distributionally robust optimization (DRO) framework with dynamic regularization to mitigate client drift. \textit{DRDM} frames the training as a min-max optimization problem aimed at maximizing performance for the worst-case client, thereby promoting robustness and fairness. This robust objective is optimized through an algorithm leveraging dynamic regularization and efficient local updates, which significantly reduces the required number of communication rounds. Moreover, we provide a theoretical convergence analysis for convex smooth objectives under partial participation. Extensive experiments on three benchmark datasets, covering various model architectures and data heterogeneity levels, demonstrate that \textit{DRDM} significantly improves worst-case test accuracy while requiring fewer communication rounds than existing state-of-the-art baselines. Furthermore, we analyze the impact of signal-to-noise ratio (SNR) and bandwidth on the energy consumption of participating clients, demonstrating that the number of local update steps can be adaptively selected to achieve a target worst-case test accuracy with minimal total energy cost across diverse communication environments.
title Distributionally Robust Federated Learning with Client Drift Minimization
topic Machine Learning
url https://arxiv.org/abs/2505.15371