Efficiently Deciding Algebraic Equivalence of Bow-Free Acyclic Path Diagrams

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: van Ommen, Thijs
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916285751754752
author van Ommen, Thijs
author_facet van Ommen, Thijs
contents For causal discovery in the presence of latent confounders, constraints beyond conditional independences exist that can enable causal discovery algorithms to distinguish more pairs of graphs. Such constraints are not well-understood yet. In the setting of linear structural equation models without bows, we study algebraic constraints and argue that these provide the most fine-grained resolution achievable. We propose efficient algorithms that decide whether two graphs impose the same algebraic constraints, or whether the constraints imposed by one graph are a subset of those imposed by another graph.
format Preprint
id arxiv_https___arxiv_org_abs_2406_09049
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Efficiently Deciding Algebraic Equivalence of Bow-Free Acyclic Path Diagrams
van Ommen, Thijs
Machine Learning
Statistics Theory
For causal discovery in the presence of latent confounders, constraints beyond conditional independences exist that can enable causal discovery algorithms to distinguish more pairs of graphs. Such constraints are not well-understood yet. In the setting of linear structural equation models without bows, we study algebraic constraints and argue that these provide the most fine-grained resolution achievable. We propose efficient algorithms that decide whether two graphs impose the same algebraic constraints, or whether the constraints imposed by one graph are a subset of those imposed by another graph.
title Efficiently Deciding Algebraic Equivalence of Bow-Free Acyclic Path Diagrams
topic Machine Learning
Statistics Theory
url https://arxiv.org/abs/2406.09049