FedGES: A Federated Learning Approach for BN Structure Learning

TL;DR

FedGES introduces a privacy-preserving federated learning method for Bayesian Network structure learning, enabling decentralized collaboration without sharing raw data, and demonstrates effectiveness on high-dimensional and sparse datasets.

Contribution

This paper presents FedGES, the first federated learning approach for BN structure learning that exchanges only network structures, not data or parameters, enhancing privacy and security.

Findings

Effective in high-dimensional settings

Works well with sparse data

Outperforms centralized methods in privacy-preserving scenarios

Abstract

Bayesian Network (BN) structure learning traditionally centralizes data, raising privacy concerns when data is distributed across multiple entities. This research introduces Federated GES (FedGES), a novel Federated Learning approach tailored for BN structure learning in decentralized settings using the Greedy Equivalence Search (GES) algorithm. FedGES uniquely addresses privacy and security challenges by exchanging only evolving network structures, not parameters or data. It realizes collaborative model development, using structural fusion to combine the limited models generated by each client in successive iterations. A controlled structural fusion is also proposed to enhance client consensus when adding any edge. Experimental results on various BNs from {\sf bnlearn}'s BN Repository validate the effectiveness of FedGES, particularly in high-dimensional (a large number of variables) and sparse data scenarios, offering a practical and privacy-preserving solution for real-world BN structure learning.