Shift Happens: Mixture of Experts based Continual Adaptation in Federated Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bhope, Rahul Atul, Jayaram, K. R., Venkateswaran, Praveen, Venkatasubramanian, Nalini
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914055340425216
author Bhope, Rahul Atul
Jayaram, K. R.
Venkateswaran, Praveen
Venkatasubramanian, Nalini
author_facet Bhope, Rahul Atul
Jayaram, K. R.
Venkateswaran, Praveen
Venkatasubramanian, Nalini
contents Federated Learning (FL) enables collaborative model training across decentralized clients without sharing raw data, yet faces significant challenges in real-world settings where client data distributions evolve dynamically over time. This paper tackles the critical problem of covariate and label shifts in streaming FL environments, where non-stationary data distributions degrade model performance and necessitate a middleware layer that adapts FL to distributional shifts. We introduce ShiftEx, a shift-aware mixture of experts framework that dynamically creates and trains specialized global models in response to detected distribution shifts using Maximum Mean Discrepancy for covariate shifts. The framework employs a latent memory mechanism for expert reuse and implements facility location-based optimization to jointly minimize covariate mismatch, expert creation costs, and label imbalance. Through theoretical analysis and comprehensive experiments on benchmark datasets, we demonstrate 5.5-12.9 percentage point accuracy improvements and 22-95 % faster adaptation compared to state-of-the-art FL baselines across diverse shift scenarios. The proposed approach offers a scalable, privacy-preserving middleware solution for FL systems operating in non-stationary, real-world conditions while minimizing communication and computational overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18789
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Shift Happens: Mixture of Experts based Continual Adaptation in Federated Learning
Bhope, Rahul Atul
Jayaram, K. R.
Venkateswaran, Praveen
Venkatasubramanian, Nalini
Machine Learning
Artificial Intelligence
Federated Learning (FL) enables collaborative model training across decentralized clients without sharing raw data, yet faces significant challenges in real-world settings where client data distributions evolve dynamically over time. This paper tackles the critical problem of covariate and label shifts in streaming FL environments, where non-stationary data distributions degrade model performance and necessitate a middleware layer that adapts FL to distributional shifts. We introduce ShiftEx, a shift-aware mixture of experts framework that dynamically creates and trains specialized global models in response to detected distribution shifts using Maximum Mean Discrepancy for covariate shifts. The framework employs a latent memory mechanism for expert reuse and implements facility location-based optimization to jointly minimize covariate mismatch, expert creation costs, and label imbalance. Through theoretical analysis and comprehensive experiments on benchmark datasets, we demonstrate 5.5-12.9 percentage point accuracy improvements and 22-95 % faster adaptation compared to state-of-the-art FL baselines across diverse shift scenarios. The proposed approach offers a scalable, privacy-preserving middleware solution for FL systems operating in non-stationary, real-world conditions while minimizing communication and computational overhead.
title Shift Happens: Mixture of Experts based Continual Adaptation in Federated Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.18789