Polyhedral aspects of score equivalence in Bayesian network structure learning

TL;DR

This paper explores the geometric structure of polytopes related to Bayesian network learning, characterizing score equivalence and its implications for optimization and constraints in integer linear programming.

Contribution

It provides a complete characterization of score equivalence faces and facets of key polytopes, linking them to supermodular functions and improving optimization strategies.

Findings

Characterization of SE objectives and faces of the polytopes.

Identification of SE facets via supermodular functions.

Guidance on constraint elimination in optimization.

Abstract

This paper deals with faces and facets of the family-variable polytope and the characteristic-imset polytope, which are special polytopes used in integer linear programming approaches to statistically learn Bayesian network structure. A common form of linear objectives to be maximized in this area leads to the concept of score equivalence (SE), both for linear objectives and for faces of the family-variable polytope. We characterize the linear space of SE objectives and establish a one-to-one correspondence between SE faces of the family-variable polytope, the faces of the characteristic-imset polytope, and standardized supermodular functions. The characterization of SE facets in terms of extremality of the corresponding supermodular function gives an elegant method to verify whether an inequality is SE-facet-defining for the family-variable polytope. We also show that when maximizing an SE objective one can eliminate linear constraints of the family-variable polytope that correspond to non-SE facets. However, we show that solely considering SE facets is not enough as a counter-example shows; one has to consider the linear inequality constraints that correspond to facets of the characteristic-imset polytope despite the fact that they may not define facets in the family-variable mode.