• English
    • español
  • English 
    • English
    • español
  • Login
View Item 
  •   Home
  • 2.- Investigación
  • Artículos
  • View Item
  •   Home
  • 2.- Investigación
  • Artículos
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

A probability estimation based criterion for model evaluation

Thumbnail
View/Open
IIT-97-033A.pdf (443.3Kb)
Date
1997-09-29
Author
Czernichow, Thomas
Muñoz San Roque, Antonio
Estado
info:eu-repo/semantics/publishedVersion
Metadata
Show full item record
Mostrar METS del ítem
Ver registro en CKH

Refworks Export

Abstract
 
 
We develop a criterion based on the estimation of the joint probability density function (pdf) of the input and the error, and on the pdf of the input. It is made to decide when the couple inputmodel no longer fit together. The estimation of the pdf is made through a Probabilistic Radial Basis Function Network (PRBFN) which can also be used to estimate the given task. We compare the results when using a dedicated network, or when extracting the density value directly from the network which estimates the input-output mapping.
 
URI
http://hdl.handle.net/11531/97842
A probability estimation based criterion for model evaluation
Tipo de Actividad
Capítulos en libros
Materias/ categorías / ODS
Instituto de Investigación Tecnológica (IIT)
Palabras Clave

Neural Networks
Collections
  • Artículos

Repositorio de la Universidad Pontificia Comillas copyright © 2015  Desarrollado con DSpace Software
Contact Us | Send Feedback
 

 

Búsqueda semántica (CKH Explorer)


Browse

All of DSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsxmlui.ArtifactBrowser.Navigation.browse_advisorxmlui.ArtifactBrowser.Navigation.browse_typeThis CollectionBy Issue DateAuthorsTitlesSubjectsxmlui.ArtifactBrowser.Navigation.browse_advisorxmlui.ArtifactBrowser.Navigation.browse_type

My Account

LoginRegister

Repositorio de la Universidad Pontificia Comillas copyright © 2015  Desarrollado con DSpace Software
Contact Us | Send Feedback