FedHarmony: Harmonizing Heterogeneous Label Correlations in Federated Multi-Label Learning

2026-04-30

AI Summary

This work addresses the challenge of label correlation drift in federated multi-label learning, which arises from heterogeneous label spaces and divergent co-occurrence patterns across clients. To mitigate local correlation bias, the authors propose a global teacher signal mechanism grounded in consensus-based label correlations. They further introduce a dynamic weighted aggregation strategy that jointly considers local data volume and correlation quality, complemented by an accelerated optimization algorithm to enhance convergence without compromising accuracy. Extensive experiments on multiple real-world federated multi-label datasets demonstrate that the proposed method significantly outperforms current state-of-the-art approaches, confirming its effectiveness and robustness.

Abstract

Federated Multi-Label Learning is a distributed paradigm where multiple clients possess heterogeneous multi-label data and perform collaborative learning under privacy constraints without sharing raw data. However, modeling label correlations under heterogeneous distributions remains challenging. Due to client-specific label spaces and varying co-occurrence patterns, correlations learned by individual clients inevitably deviate from the global structure, a phenomenon we term label correlation drift. To address this, we propose FedHarmony, a framework that harmonizes heterogeneous label correlations across clients. It introduces consensus correlation, capturing agreement among other clients and serving as a global teacher to correct biased local estimates. During aggregation, FedHarmony evaluates each client by both data size and correlation quality, assigning weights accordingly. Moreover, we develop an accelerated optimization algorithm for FedHarmony and theoretically establish faster convergence without sacrificing accuracy. Experiments on real-world federated multi-label datasets show that FedHarmony consistently outperforms state-of-the-art methods.

Problem

• Federated Multi-Label Learning
• Label Correlation Drift
• Heterogeneous Label Spaces
• Label Co-occurrence Patterns
• Privacy-Preserving Collaboration

Innovation

• federated multi-label learning
• label correlation drift
• consensus correlation
• heterogeneous data
• accelerated optimization