FedMSGL: A Self-Expressive Hypergraph Based Federated Multi-View Learning

2025-03-12

AI Summary

In federated learning, heterogeneous feature dimensions cause global model bias toward high-dimensional clients, while single-view modeling fails to capture structural correlations across multi-view data. To address these challenges, we propose FedMSGL—the first federated multi-view learning framework integrating self-expressiveness with hypergraph structure. Its core innovations include: (i) constructing a cross-view, cross-dimensional self-expressive hypergraph to jointly learn a unified subspace and enable adaptive view fusion; and (ii) designing a federated multi-view subspace alignment mechanism coupled with a dimension-aware weighted aggregation strategy to mitigate dimensional bias. Extensive experiments on heterogeneous multi-view datasets demonstrate that FedMSGL achieves an average 3.2% improvement in classification accuracy, accelerates convergence by 21%, and reduces communication overhead by 18% compared to state-of-the-art baselines.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Multiagent Systems: Multiagent Learning

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Federated learning is essential for enabling collaborative model training across decentralized data sources while preserving data privacy and security. This approach mitigates the risks associated with centralized data collection and addresses concerns related to data ownership and compliance. Despite significant advancements in federated learning algorithms that address communication bottlenecks and enhance privacy protection, existing works overlook the impact of differences in data feature dimensions, resulting in global models that disproportionately depend on participants with large feature dimensions. Additionally, current single-view federated learning methods fail to account for the unique characteristics of multi-view data, leading to suboptimal performance in processing such data. To address these issues, we propose a Self-expressive Hypergraph Based Federated Multi-view Learning method (FedMSGL). The proposed method leverages self-expressive character in the local training to learn uniform dimension subspace with latent sample relation. At the central side, an adaptive fusion technique is employed to generate the global model, while constructing a hypergraph from the learned global and view-specific subspace to capture intricate interconnections across views. Experiments on multi-view datasets with different feature dimensions validated the effectiveness of the proposed method.

Problem

• Addresses feature dimension disparities in federated learning.
• Enhances multi-view data processing in federated learning.
• Improves global model accuracy with adaptive fusion techniques.

Innovation

• Self-expressive hypergraph for multi-view learning
• Adaptive fusion technique for global model generation
• Uniform dimension subspace learning in local training