Learning Bayesian network classifiers with completed partially directed acyclic graphs

Mihaljevic, Bojan and Bielza Lozoya, María Concepción and Larrañaga Múgica, Pedro María (2018). Learning Bayesian network classifiers with completed partially directed acyclic graphs. In: "International Conference on Probabilistic Graphical Models", 11(09/2018-14/09/2018, Praga, República Checa. ISBN 2640-3498. pp. 272-283.

Description

Title: Learning Bayesian network classifiers with completed partially directed acyclic graphs
Author/s:
  • Mihaljevic, Bojan
  • Bielza Lozoya, María Concepción
  • Larrañaga Múgica, Pedro María
Item Type: Presentation at Congress or Conference (Article)
Event Title: International Conference on Probabilistic Graphical Models
Event Dates: 11(09/2018-14/09/2018
Event Location: Praga, República Checa
Title of Book: Proceedings of Machine Learning Research
Date: 2018
ISBN: 2640-3498
Volume: 72
Subjects:
Freetext Keywords: equivalence class; greedy equivalence search; augmented naive Bayes
Faculty: E.T.S.I. de Sistemas 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 (322kB) | Preview

Abstract

Most search and score algorithms for learning Bayesian network classifiers from data traverse the space of directed acyclic graphs (DAGs), making arbitrary yet possibly suboptimal arc directionality decisions. This can be remedied by learning in the space of DAG equivalence classes. We provide a number of contributions to existing work along this line. First, we identify the smallest subspace of DAGs that covers all possible class-posterior distributions when data is complete. All the DAGs in this space, which we call minimal class-focused DAGs (MC-DAGs), are such that their every arc is directed towards a child of the class variable. Second, in order to traverse the equivalence classes of MC-DAGs, we adapt the greedy equivalence search (GES) by adding operator validity criteria which ensure GES only visits states within our space. Third, we specify how to efficiently evaluate the discriminative score of a GES operator for MC-DAG in time independent of the number of variables and without converting the completed partially DAG, which represents an equivalence class, into a DAG. The adapted GES perfomed well on real-world data sets.

Funding Projects

TypeCodeAcronymLeaderTitle
Horizon 2020785907HBP SGA2UnspecifiedHuman Brain Project Specific Grant Agreement 2
Government of SpainC080020-09UnspecifiedUnspecifiedCajal Blue Brain
Government of SpainTIN2016-79684-PUnspecifiedUnspecifiedUnspecified
Madrid Regional GovernmentS2013/ICE-2845CASI – CAMUnspecifiedConceptos y aplicaciones de los sistemas inteligentes

More information

Item ID: 54651
DC Identifier: http://oa.upm.es/54651/
OAI Identifier: oai:oa.upm.es:54651
Official URL: http://proceedings.mlr.press/v72/
Deposited by: Memoria Investigacion
Deposited on: 23 Jan 2020 15:34
Last Modified: 24 Jan 2020 11:01
  • 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