Improving Federated Relational Data Modeling via Basis Alignment and Weight Penalty
This work tackles the problem of applying federated learning to relational data (Knowledge Graphs) for researchers and practitioners in privacy-preserving collaborative learning, representing an incremental advancement in FL methodologies.
This paper addresses the challenge of federated learning on relational data, specifically Knowledge Graphs (KGs), which has seen limited research. The authors propose FedAlign, a modified graph neural network algorithm that incorporates optimal transportation for on-client personalization and a weight constraint to accelerate convergence, outperforming state-of-the-art FL methods like FedAVG and FedProx with better convergence.
Federated learning (FL) has attracted increasing attention in recent years. As a privacy-preserving collaborative learning paradigm, it enables a broader range of applications, especially for computer vision and natural language processing tasks. However, to date, there is limited research of federated learning on relational data, namely Knowledge Graph (KG). In this work, we present a modified version of the graph neural network algorithm that performs federated modeling over KGs across different participants. Specifically, to tackle the inherent data heterogeneity issue and inefficiency in algorithm convergence, we propose a novel optimization algorithm, named FedAlign, with 1) optimal transportation (OT) for on-client personalization and 2) weight constraint to speed up the convergence. Extensive experiments have been conducted on several widely used datasets. Empirical results show that our proposed method outperforms the state-of-the-art FL methods, such as FedAVG and FedProx, with better convergence.