Learning tractable multidimensional Bayesian network classifiers

Benjumeda Barquita, Marco Alberto; Larrañaga Múgica, Pedro María y Bielza Lozoya, María Concepción (2016). Learning tractable multidimensional Bayesian network classifiers. En: "Eighth International Conference on Probabilistic Graphical Models", 06-09 Sep 2016, Lugano. pp. 13-24.

Descripción

Título: Learning tractable multidimensional Bayesian network classifiers
Autor/es:
  • Benjumeda Barquita, Marco Alberto
  • Larrañaga Múgica, Pedro María
  • Bielza Lozoya, María Concepción
Tipo de Documento: Ponencia en Congreso o Jornada (Artículo)
Título del Evento: Eighth International Conference on Probabilistic Graphical Models
Fechas del Evento: 06-09 Sep 2016
Lugar del Evento: Lugano
Título del Libro: Proceedings of the Eighth International Conference on Probabilistic Graphical Models
Fecha: 2016
Volumen: 52
Materias:
Palabras Clave Informales: Multidimensional classification; Bayesian network classifiers; MPE complexity; learning from data
Escuela: E.T.S. de Ingenieros Informáticos (UPM)
Departamento: Inteligencia Artificial
Licencias Creative Commons: Reconocimiento - Sin obra derivada - No comercial

Texto completo

[img]
Vista Previa
PDF (Document Portable Format) - Se necesita un visor de ficheros PDF, como GSview, Xpdf o Adobe Acrobat Reader
Descargar (214kB) | Vista Previa

Resumen

Multidimensional classification has become one of the most relevant topics in view of the many domains that require a vector of class values to be assigned to a vector of given features. The popularity of multidimensional Bayesian network classifiers has increased in the last few years due to their expressive power and the existence of methods for learning different families of these models. The problem with this approach is that the computational cost of using the learned models is usually high, especially if there are a lot of class variables. Class-bridge decomposability means that the multidimensional classification problem can be divided into multiple subproblems for these models. In this paper, we prove that class-bridge decomposability can also be used to guarantee the tractability of the models. We also propose a strategy for efficiently bounding their inference complexity, providing a simple learning method with an order-based search that obtains tractable multidimensional Bayesian network classifiers. Experimental results show that our approach is competitive with other methods in the state of the art and ensures the tractability of the learned models.

Proyectos asociados

TipoCódigoAcrónimoResponsableTítulo
Gobierno de EspañaC080020-09Sin especificarSin especificarCajal Blue Brain Proyect
Comunidad de MadridTIN2013-41592S2013/ICE-2845-CASI-CAM-CMSin especificarCajal Blue Brain Proyect
FP7FP7/2007-2013Sin especificarSin especificarHuman Brain Project

Más información

ID de Registro: 46644
Identificador DC: http://oa.upm.es/46644/
Identificador OAI: oai:oa.upm.es:46644
Depositado por: Memoria Investigacion
Depositado el: 18 Oct 2017 11:42
Ultima Modificación: 18 Oct 2017 11:42
  • Open Access
  • Open Access
  • Sherpa-Romeo
    Compruebe si la revista anglosajona en la que ha publicado un artículo permite también su publicación en abierto.
  • Dulcinea
    Compruebe si la revista española en la que ha publicado un artículo permite también su publicación en abierto.
  • Recolecta
  • e-ciencia
  • Observatorio I+D+i UPM
  • OpenCourseWare UPM