Algebraic geometry of Gaussian Bayesian networks

DOI10.1016/J.AAM.2007.04.004zbMATH Open1175.13014arXiv0704.0918OpenAlexW2133391651MaRDI QIDQ1031738FDOQ1031738

Authors: Seth Sullivant 

Publication date: 30 October 2009

Published in: Advances in Applied Mathematics (Search for Journal in Brave)

Abstract: Conditional independence models in the Gaussian case are algebraic varieties in the cone of positive definite covariance matrices. We study these varieties in the case of Bayesian networks, with a view towards generalizing the recursive factorization theorem to situations with hidden variables. In the case when the underlying graph is a tree, we show that the vanishing ideal of the model is generated by the conditional independence statements implied by graph. We also show that the ideal of any Bayesian network is homogeneous with respect to a multigrading induced by a collection of upstream random variables. This has a number of important consequences for hidden variable models. Finally, we relate the ideals of Bayesian networks to a number of classical constructions in algebraic geometry including toric degenerations of the Grassmannian, matrix Schubert varieties, and secant varieties.

Full work available at URL: https://arxiv.org/abs/0704.0918

Recommendations

• Algebraic geometry of Bayesian networks
• Geometry, moments and conditional independence trees with hidden variables.
• Gaussian graphical models with toric vanishing ideals
• Bayesian networks for discrete multivariate data: An algebraic approach to inference
• Conditionals, Information, and Inference

zbMATH Keywords

Bayesian networkalgebraic statisticsgraphical modelmultivariate GaussianGröbner basis

Mathematics Subject Classification ID

Multivariate analysis (62H99) Gröbner bases; other bases for ideals and modules (e.g., Janet and border bases) (13P10) Applications of commutative algebra (e.g., to statistics, control theory, optimization, etc.) (13P25) Determinantal varieties (14M12)

Cites Work

• Causation, prediction, and search
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Binomial determinants, paths, and hook length formulae
• Gröbner geometry of Schubert polynomials
• Title not available (Why is that?)
• The tropical Grassmannian
• Rings of Invariants of Tori, Cohen-Macaulay Rings Generated by Monomials, and Polytopes
• Combinatorial secant varieties
• Algebraic factor analysis: tetrads, pentads and beyond
• On the ideals of secant varieties of Segre varieties
• Algebraic geometry of Bayesian networks

Cited In (24)

• Matrix Schubert varieties and Gaussian conditional independence models
• Combinatorial and algebraic perspectives on the marginal independence structure of Bayesian networks
• Initial ideals of Pfaffian ideals
• Title not available (Why is that?)
• Granger causality from quantized measurements
• Algebraic structures in statistical methodology. Abstracts from the workshop held December 4--10, 2022
• Algebraic geometry of Bayesian networks
• Directed Gaussian graphical models with toric vanishing ideals
• VC dimension and inner product space induced by Bayesian networks
• Inner product space and concept classes induced by Bayesian networks
• Geometry, moments and conditional independence trees with hidden variables.
• Logarithmic Voronoi cells for Gaussian models
• Equations defining probability tree models
• Gaussian graphical models with toric vanishing ideals
• The Geometry of Causal Probability Trees that are Algebraically Constrained
• A Coordinate System for Gaussian Networks
• Algebraic properties of HTC-identifiable graphs
• Multivariate Gaussians, semidefinite matrix completion, and convex algebraic geometry
• Conditionals, Information, and Inference
• Conjunctive Bayesian networks
• Conditional independence ideals with hidden variables
• Automorphism groups of Gaussian Bayesian networks
• A short history of algebraic statistics
• Trek separation for Gaussian graphical models

Uses Software

• Macaulay2
• TETRAD

This page was built for publication: Algebraic geometry of Gaussian Bayesian networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1031738)