On generating random Gaussian graphical models

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Standard

On generating random Gaussian graphical models. / Córdoba, Irene; Varando, Gherardo; Bielza, Concha; Larrañaga, Pedro.

I: International Journal of Approximate Reasoning, Bind 125, 2020, s. 240-250.

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Harvard

Córdoba, I, Varando, G, Bielza, C & Larrañaga, P 2020, 'On generating random Gaussian graphical models', International Journal of Approximate Reasoning, bind 125, s. 240-250. https://doi.org/10.1016/j.ijar.2020.07.007

APA

Córdoba, I., Varando, G., Bielza, C., & Larrañaga, P. (2020). On generating random Gaussian graphical models. International Journal of Approximate Reasoning, 125, 240-250. https://doi.org/10.1016/j.ijar.2020.07.007

Vancouver

Córdoba I, Varando G, Bielza C, Larrañaga P. On generating random Gaussian graphical models. International Journal of Approximate Reasoning. 2020;125:240-250. https://doi.org/10.1016/j.ijar.2020.07.007

Author

Córdoba, Irene ; Varando, Gherardo ; Bielza, Concha ; Larrañaga, Pedro. / On generating random Gaussian graphical models. I: International Journal of Approximate Reasoning. 2020 ; Bind 125. s. 240-250.

Bibtex

@article{ef34b1a46aa640df8649112ff3f3337e,
title = "On generating random Gaussian graphical models",
abstract = "Structure learning methods for covariance and concentration graphs are often validated on synthetic models, usually obtained by randomly generating: (i) an undirected graph, and (ii) a compatible symmetric positive definite (SPD) matrix. In order to ensure positive definiteness in (ii), a dominant diagonal is usually imposed. In this work we investigate different methods to generate random symmetric positive definite matrices with undirected graphical constraints. We show that if the graph is chordal it is possible to sample uniformly from the set of correlation matrices compatible with the graph, while for general undirected graphs we rely on a partial orthogonalization method.",
keywords = "Algorithm validation, Concentration graph, Covariance graph, Positive definite matrix simulation, Undirected graphical model",
author = "Irene C{\'o}rdoba and Gherardo Varando and Concha Bielza and Pedro Larra{\~n}aga",
year = "2020",
doi = "10.1016/j.ijar.2020.07.007",
language = "English",
volume = "125",
pages = "240--250",
journal = "International Journal of Approximate Reasoning",
issn = "0888-613X",
publisher = "Elsevier",

}

RIS

TY - JOUR

T1 - On generating random Gaussian graphical models

AU - Córdoba, Irene

AU - Varando, Gherardo

AU - Bielza, Concha

AU - Larrañaga, Pedro

PY - 2020

Y1 - 2020

N2 - Structure learning methods for covariance and concentration graphs are often validated on synthetic models, usually obtained by randomly generating: (i) an undirected graph, and (ii) a compatible symmetric positive definite (SPD) matrix. In order to ensure positive definiteness in (ii), a dominant diagonal is usually imposed. In this work we investigate different methods to generate random symmetric positive definite matrices with undirected graphical constraints. We show that if the graph is chordal it is possible to sample uniformly from the set of correlation matrices compatible with the graph, while for general undirected graphs we rely on a partial orthogonalization method.

AB - Structure learning methods for covariance and concentration graphs are often validated on synthetic models, usually obtained by randomly generating: (i) an undirected graph, and (ii) a compatible symmetric positive definite (SPD) matrix. In order to ensure positive definiteness in (ii), a dominant diagonal is usually imposed. In this work we investigate different methods to generate random symmetric positive definite matrices with undirected graphical constraints. We show that if the graph is chordal it is possible to sample uniformly from the set of correlation matrices compatible with the graph, while for general undirected graphs we rely on a partial orthogonalization method.

KW - Algorithm validation

KW - Concentration graph

KW - Covariance graph

KW - Positive definite matrix simulation

KW - Undirected graphical model

UR - http://www.scopus.com/inward/record.url?scp=85089486344&partnerID=8YFLogxK

U2 - 10.1016/j.ijar.2020.07.007

DO - 10.1016/j.ijar.2020.07.007

M3 - Journal article

AN - SCOPUS:85089486344

VL - 125

SP - 240

EP - 250

JO - International Journal of Approximate Reasoning

JF - International Journal of Approximate Reasoning

SN - 0888-613X

ER -

ID: 248192670