Markov Blanket Discovery using Minimum Message Length

TL;DR

This paper introduces three new methods for Markov Blanket discovery using Minimum Message Length, demonstrating their effectiveness and scalability for causal Bayesian network learning from large datasets.

Contribution

The paper develops and empirically evaluates three novel MML-based Markov Blanket discovery methods, improving scalability and performance in causal discovery tasks.

Findings

MML-based methods are competitive with existing approaches.

The best MML method shows advantageous features in scalability.

Empirical results favor the proposed methods over some traditional techniques.

Abstract

Causal discovery automates the learning of causal Bayesian networks from data and has been of active interest from their beginning. With the sourcing of large data sets off the internet, interest in scaling up to very large data sets has grown. One approach to this is to parallelize search using Markov Blanket (MB) discovery as a first step, followed by a process of combining MBs in a global causal model. We develop and explore three new methods of MB discovery using Minimum Message Length (MML) and compare them empirically to the best existing methods, whether developed specifically as MB discovery or as feature selection. Our best MML method is consistently competitive and has some advantageous features.