Learning All Credible Bayesian Network Structures for Model Averaging

TL;DR

This paper introduces a scalable method for Bayesian network model averaging that considers only credible models, improving accuracy and efficiency over previous approaches, especially for larger networks.

Contribution

The paper presents a novel, scalable approach to Bayesian network model averaging that focuses on credible models and outperforms existing methods in size and efficiency.

Findings

Scales to larger Bayesian networks than previous methods

Focuses on near-optimal models for better accuracy

Improves efficiency in model averaging process

Abstract

A Bayesian network is a widely used probabilistic graphical model with applications in knowledge discovery and prediction. Learning a Bayesian network (BN) from data can be cast as an optimization problem using the well-known score-and-search approach. However, selecting a single model (i.e., the best scoring BN) can be misleading or may not achieve the best possible accuracy. An alternative to committing to a single model is to perform some form of Bayesian or frequentist model averaging, where the space of possible BNs is sampled or enumerated in some fashion. Unfortunately, existing approaches for model averaging either severely restrict the structure of the Bayesian network or have only been shown to scale to networks with fewer than 30 random variables. In this paper, we propose a novel approach to model averaging inspired by performance guarantees in approximation algorithms. Our approach has two primary advantages. First, our approach only considers credible models in that they are optimal or near-optimal in score. Second, our approach is more efficient and scales to significantly larger Bayesian networks than existing approaches.