FedMSGL: A Self-Expressive Hypergraph Based Federated Multi-View Learning
Daoyuan Li, Zuyuan Yang, Shengli Xie
TL;DR
This work tackles privacy-preserving federated learning with two key gaps: (i) disparate feature dimensions across participants and (ii) the lack of multi-view handling in vertical FL. The authors introduce FedMSGL, which performs local self-expressive subspace learning to obtain a uniform-dimensional representation split into a global-consistent part $\mathbf{C}^k$ and a view-specific part $\mathbf{U}^k$, followed by a central adaptive fusion to form a global subspace $\mathbf{G}$ and a hypergraph-based integration via the Laplacian $\mathbf{L}_h$ guided clustering with indicator $\mathbf{F}$. The method alternates between local updates of $\mathbf{C}^k,\mathbf{U}^k$ and global updates of $\mathbf{G},\mathbf{F}$, enabling privacy-preserving cross-view collaboration. Empirical results on five multi-view datasets show FedMSGL achieves competitive performance with centralized MVL methods and outperforms existing federated MVL baselines, demonstrating effective handling of multi-view data and feature-dimension heterogeneity in federated settings.
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.