On a hypergraph probabilistic graphical model

DOI10.1007/S10472-020-09701-7zbMATH Open1465.62112arXiv1811.08372OpenAlexW3040946540MaRDI QIDQ2023886FDOQ2023886

Authors: Mohammad Ali Javidian, Zhiyu Wang, Linyuan Lu, Marco Valtorta 

Publication date: 3 May 2021

Published in: Annals of Mathematics and Artificial Intelligence (Search for Journal in Brave)

Abstract: We propose a directed acyclic hypergraph framework for a probabilistic graphical model that we call Bayesian hypergraphs. The space of directed acyclic hypergraphs is much larger than the space of chain graphs. Hence Bayesian hypergraphs can model much finer factorizations than Bayesian networks or LWF chain graphs and provide simpler and more computationally efficient procedures for factorizations and interventions. Bayesian hypergraphs also allow a modeler to represent causal patterns of interaction such as Noisy-OR graphically (without additional annotations). We introduce global, local and pairwise Markov properties of Bayesian hypergraphs and prove under which conditions they are equivalent. We define a projection operator, called shadow, that maps Bayesian hypergraphs to chain graphs, and show that the Markov properties of a Bayesian hypergraph are equivalent to those of its corresponding chain graph. We extend the causal interpretation of LWF chain graphs to Bayesian hypergraphs and provide corresponding formulas and a graphical criterion for intervention.

Full work available at URL: https://arxiv.org/abs/1811.08372

Recommendations

• Graphs for margins of Bayesian networks
• Chain Graph Models and their Causal Interpretations
• scientific article; zbMATH DE number 1269592
• Causal inference from graphical models
• scientific article; zbMATH DE number 956594

zbMATH Keywords

structural equation modelBayesian networkgraphical modelfactorizationhypergraphinterventionchain graphMarkov property

Mathematics Subject Classification ID

Probabilistic graphical models (62H22) Graphical methods in statistics (62A09) Hypergraphs (05C65) Edge subsets with special properties (factorization, matching, partitioning, covering and packing, etc.) (05C70) Structural characterization of families of graphs (05C75)

Cites Work

• Causation, prediction, and search
• Graphical models for associations between variables, some of which are qualitative and some quantitative
• Causal inference from graphical models
• Title not available (Why is that?)
• Causality. Models, reasoning, and inference
• Ancestral graph Markov models.
• Title not available (Why is that?)
• Title not available (Why is that?)
• Conditional independence for statistical operations
• Linear dependencies represented by chain graphs. With comments and a rejoinder by the authors
• Bayesian Networks and Decision Graphs
• Causal diagrams for empirical research
• Markov Properties for Acyclic Directed Mixed Graphs
• Chain Graph Models and their Causal Interpretations
• Local computation with valuations from a commutative semigroup
• Title not available (Why is that?)
• Title not available (Why is that?)
• On the Validity of the Markov Interpretation of Path Diagrams of Gaussian Structural Equations Systems with Correlated Errors
• Markov fields and log-linear interaction models for contingency tables
• Marginal AMP chain graphs
• Alternative Markov properties for chain graphs
• Title not available (Why is that?)
• Factor graphs and the sum-product algorithm
• Discrete chain graph models
• Joint Response Graphs and Separation Induced by Triangular Systems
• Independence properties of directed markov fields
• Graphical and Recursive Models for Contingency Tables
• An Introduction to Spatial Point Processes and Markov Random Fields
• Two operations of merging and splitting components in a chain graph
• Title not available (Why is that?)

Cited In (2)

• Factor metanetwork: a multilevel probabilistic meta-model based on factor graphs
• Graphs for margins of Bayesian networks

Uses Software

• TETRAD

This page was built for publication: On a hypergraph probabilistic graphical model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2023886)