FedRD: Reducing Divergences for Generalized Federated Learning via Heterogeneity-aware Parameter Guidance

2026-01-28

AI Summary

This work addresses the challenge of generalization to unseen clients in heterogeneous federated learning by explicitly identifying and jointly tackling two key issues: optimization divergence and performance divergence. To this end, the authors propose FedRD, a novel algorithm that employs a heterogeneity-aware parameter guidance mechanism to jointly optimize global model aggregation and local debiased classifier training. This approach effectively narrows the performance gap between participating and unseen clients. Extensive experiments on multiple public multi-domain datasets demonstrate that FedRD significantly outperforms existing methods, substantially enhancing the model’s generalization capability to newly joined, heterogeneous clients.

Technology Category

Machine Learning: Distributed Machine Learning & Federated LearningSearch and Optimization: Learning to SearchComputer Vision: Learning & Optimization for CV

Application Category

User Modeling, Personalization and Recommendation: Federated recommendation systems and personalizationSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingResponsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Heterogeneous federated learning (HFL) aims to ensure effective and privacy-preserving collaboration among different entities. As newly joined clients require significant adjustments and additional training to align with the existing system, the problem of generalizing federated learning models to unseen clients under heterogeneous data has become progressively crucial. Consequently, we highlight two unsolved challenging issues in federated domain generalization: Optimization Divergence and Performance Divergence. To tackle the above challenges, we propose FedRD, a novel heterogeneity-aware federated learning algorithm that collaboratively utilizes parameter-guided global generalization aggregation and local debiased classification to reduce divergences, aiming to obtain an optimal global model for participating and unseen clients. Extensive experiments on public multi-domain datasets demonstrate that our approach exhibits a substantial performance advantage over competing baselines in addressing this specific problem.

Problem

• Heterogeneous Federated Learning
• Federated Domain Generalization
• Optimization Divergence
• Performance Divergence
• Unseen Clients

Innovation

• Federated Domain Generalization
• Heterogeneous Federated Learning
• Parameter-guided Aggregation
• Optimization Divergence
• Local Debiasing