• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.55.2018.tde-06032018-104226
Document
Author
Full name
Estéfane George Macedo de Lacerda
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 1999
Supervisor
Committee
Carvalho, André Carlos Ponce de Leon Ferreira de (President)
Braga, Antonio de Padua
Rezende, Solange Oliveira
Title in Portuguese
Otimização de Redes Neurais RBF Usando Algoritmos Genéticos e sua Aplicação na Área Financeira"
Keywords in Portuguese
Não disponível
Abstract in Portuguese
A escolha da topologia de uma Rede Neural RBF é geralmente realizada por tentativa e erro baseado na experiência do projetista. Os algoritmos de treinamento existentes que determinam a topologia da rede utilizam métodos locais, que apresentam uma grande possibilidade de cair em mínimos locais gerando soluções sub-ótimas. Algoritmos Genéticos representam um método de busca global apropriado para encontrar boas soluções em espaços de busca complexos, como o espaço de busca das topologias das Redes Neurais. Este trabalho propõe um Algoritmo Genético para otimizar a topologia de redes RBF limitando o espaço de busca através de uma técnica de aglomeração. Os resultados obtidos sugerem que esta otimização melhora o desempenho de redes RBF em aplicações financeiras.
Title in English
Not available
Keywords in English
Not available
Abstract in English
The choice of the topology of a RBF Neural Network is usually carried out by trial and error based on the designer experience. The most common training algorithms that define the network topology use local methods which have a large possibility of being trapped at a local minima, producing sub-optima solutions. Genetic Algorithms represent a global search method appropriate to find good solutions in complex search spaces, like the space of Neural Networks topologies. This work proposes a Genetic Algorithm for RBF networks optimisation limiting the search space through a clustering technique. The results achieved suggest that this optimisation improves the performance of RBF networks in finance applications.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2018-03-19
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.