A PC Algorithm for Max-Linear Bayesian Networks

TL;DR

This paper introduces a new PC-based algorithm called PCstar for discovering the structure of Max-Linear Bayesian Networks, which are challenging for traditional methods due to their unique independence properties.

Contribution

The paper develops a novel constraint-based causal discovery algorithm tailored for MLBNs, extending the PC algorithm to handle their non-faithfulness and unique separation criteria.

Findings

PC remains consistent with $ ext{*}$-separation oracle.

PCstar can orient additional edges beyond traditional methods.

The approach enables accurate structure learning in heavy-tailed distributions.

Abstract

Max-linear Bayesian networks (MLBNs) are a relatively recent class of structural equation models which arise when the random variables involved have heavy-tailed distributions. Unlike most directed graphical models, MLBNs are typically not faithful to d-separation and thus classical causal discovery algorithms such as the PC algorithm or greedy equivalence search can not be used to accurately recover the true graph structure. In this paper, we begin the study of constraint-based discovery algorithms for MLBNs given an oracle for testing conditional independence in the true, unknown graph. We show that if the oracle is given by the $\ast$-separation criteria in the true graph, then the PC algorithm remains consistent despite the presence of additional CI statements implied by $\ast$-separation. We also introduce a new causal discovery algorithm named "PCstar" which assumes faithfulness to $C^\ast$-separation and is able to orient additional edges which cannot be oriented with only d- or $\ast$-separation.