Improving network generalization through selection of examples

In this work, we study how the selection of examples affects the learning procedure in a neural network and its relationship with the complexity of the function under study and its architecture. We focus on three different problems: parity, addition of two number and bitshifting implemented on feed...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Cannas, Sergio A., Franco, Leonardo
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 1998
Materias:
Ciencias Informáticas
Informática
Neural nets
Network Architecture and Design
neural networks
machine learning
generalization
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/24820
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-24820
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Informática
Neural nets
Network Architecture and Design
neural networks
machine learning
generalization
spellingShingle Ciencias Informáticas
Informática
Neural nets
Network Architecture and Design
neural networks
machine learning
generalization
Cannas, Sergio A.
Franco, Leonardo
Improving network generalization through selection of examples
topic_facet Ciencias Informáticas
Informática
Neural nets
Network Architecture and Design
neural networks
machine learning
generalization
description In this work, we study how the selection of examples affects the learning procedure in a neural network and its relationship with the complexity of the function under study and its architecture. We focus on three different problems: parity, addition of two number and bitshifting implemented on feed-forward Neural Networks. For the parity problem, one of the most used problems for testing learning algorithms, we obtain the result that only the use of the whole set of examples assures global learnings. For the other two functions we show that generalization can be considerably improved with a particular selection of examples instead of a random one.
format Objeto de conferencia
Objeto de conferencia
author Cannas, Sergio A.
Franco, Leonardo
author_facet Cannas, Sergio A.
Franco, Leonardo
author_sort Cannas, Sergio A.
title Improving network generalization through selection of examples
title_short Improving network generalization through selection of examples
title_full Improving network generalization through selection of examples
title_fullStr Improving network generalization through selection of examples
title_full_unstemmed Improving network generalization through selection of examples
title_sort improving network generalization through selection of examples
publishDate 1998
url http://sedici.unlp.edu.ar/handle/10915/24820
work_keys_str_mv AT cannassergioa improvingnetworkgeneralizationthroughselectionofexamples
AT francoleonardo improvingnetworkgeneralizationthroughselectionofexamples
bdutipo_str Repositorios
_version_ 1764820466366676993

Ejemplares similares