Bayesian optimization of the PC algorithm for learning Gaussian Bayesian networks

Córdoba Sánchez, Irene and Garrido Merchán, Eduardo César and Hernández Lobato, Daniel and Bielza Lozoya, María Concepción and Larrañaga Múgica, Pedro María (2018). Bayesian optimization of the PC algorithm for learning Gaussian Bayesian networks. In: "18th Conference of the Spanish Association for Artificial Intelligence (CAEPIA 2018)", 23-26 Oct 2018, Granada, España. ISBN 978-3-030-00373-9. pp. 44-54. https://doi.org/10.1007/978-3-030-00374-6_5.

Description

Title: Bayesian optimization of the PC algorithm for learning Gaussian Bayesian networks
Author/s:
  • Córdoba Sánchez, Irene
  • Garrido Merchán, Eduardo César
  • Hernández Lobato, Daniel
  • Bielza Lozoya, María Concepción
  • Larrañaga Múgica, Pedro María
Item Type: Presentation at Congress or Conference (Article)
Event Title: 18th Conference of the Spanish Association for Artificial Intelligence (CAEPIA 2018)
Event Dates: 23-26 Oct 2018
Event Location: Granada, España
Title of Book: Advances in Artificial Intelligence
Date: 2018
ISBN: 978-3-030-00373-9
Subjects:
Faculty: E.T.S. de Ingenieros Informáticos (UPM)
Department: Inteligencia Artificial
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (845kB) | Preview

Abstract

The PC algorithm is a popular method for learning the structure of Gaussian Bayesian networks. It carries out statistical tests to determine absent edges in the network. It is hence governed by two parameters: (i) The type of test, and (ii) its significance level. These parameters are usually set to values recommended by an expert. Nevertheless, such an approach can suffer from human bias, leading to suboptimal reconstruction results. In this paper we consider a more principled approach for choosing these parameters in an automatic way. For this we optimize a reconstruction score evaluated on a set of different Gaussian Bayesian networks. This objective is expensive to evaluate and lacks a closed-form expression, which means that Bayesian optimization (BO) is a natural choice. BO methods use a model to guide the search and are hence able to exploit smoothness properties of the objective surface. We show that the parameters found by a BO method outperform those found by a random search strategy and the expert recommendation. Importantly, we have found that an often overlooked statistical test provides the best over-all reconstruction results.

Funding Projects

TypeCodeAcronymLeaderTitle
Madrid Regional GovernmentS2013/ICE-2845CASI – CAMUnspecifiedConceptos y aplicaciones de los sistemas inteligentes
Government of SpainTIN2016-79684-PUnspecifiedUniversidad Politécnica de MadridAvances en clasificación multidimensional y detección de anomalías con redes bayesianas
Government of SpainTIN2016-76406-PUnspecifiedUniversidad Autónoma de MadridFronteras en aprendizaje automático y aplicaciones multidisciplinares
Government of SpainTEC2016-81900-REDTUnspecifiedUniversidad de ValenciaAvances en métodos núcleo para datos estructurados
Government of SpainC080020-09UnspecifiedUnspecifiedCajal Blue Brain project

More information

Item ID: 54649
DC Identifier: http://oa.upm.es/54649/
OAI Identifier: oai:oa.upm.es:54649
DOI: 10.1007/978-3-030-00374-6_5
Official URL: https://link.springer.com/content/pdf/10.1007%2F978-3-030-00374-6.pdf
Deposited by: Memoria Investigacion
Deposited on: 29 Apr 2019 10:30
Last Modified: 29 Apr 2019 10:30
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM