The Complexity of Morality: Checking Markov Blanket Consistency with DAGs via Morality

TL;DR

This paper explores the relationship between Markov blankets and moral graphs in Bayesian networks, introducing new concepts and analyzing the computational complexity of morality decision problems.

Contribution

It establishes a bijection between consistent Markov blankets and moral graphs, and introduces weak recursive simpliciality and perfect elimination kits as new concepts.

Findings

Morality can be decided in polynomial time for graphs with max degree < 5.

Deciding morality is NP-complete for graphs with higher maximum degrees.

Introduces new concepts linking Markov blankets and moral graphs.

Abstract

A family of Markov blankets in a faithful Bayesian network satisfies the symmetry and consistency properties. In this paper, we draw a bijection between families of consistent Markov blankets and moral graphs. We define the new concepts of weak recursive simpliciality and perfect elimination kits. We prove that they are equivalent to graph morality. In addition, we prove that morality can be decided in polynomial time for graphs with maximum degree less than $5$, but the problem is NP-complete for graphs with higher maximum degrees.