Logo do repositório
Tudo no RIUFPA
Documentos
Contato
Sobre
Ajuda
  • Português do Brasil
  • English
  • Español
  • Français
Entrar
Novo usuário? Clique aqui para cadastrar. Esqueceu sua senha?
  1. Início
  2. Pesquisar por Assunto

Navegando por Assunto "Algoritmos neuroevolutivos"

Filtrar resultados informando as primeiras letras
Agora exibindo 1 - 1 de 1
  • Resultados por página
  • Opções de Ordenação
  • Carregando...
    Imagem de Miniatura
    ItemAcesso aberto (Open Access)
    Uma metodologia biologicamente inspirada para projeto automático de redes neurais artificiais usando Sistemas-L paramétricos com memória
    (Universidade Federal do Pará, 2016-08-26) CAMPOS, Lidio Mauro Lima de; ROISENBERG, Mauro; http://lattes.cnpq.br/5872119613051645; OLIVEIRA, Roberto Célio Limão de; http://lattes.cnpq.br/4497607460894318
    This thesis proposes a hybrid neuro-evolutive algorithm (NEA) that uses a compact indirect encoding scheme (IES) for representing its genotypes (a set of ten production rules of a Lindenmayer System with memory), moreover has the ability to reuse the genotypes and automatically build modular, hierarchical and recurrent neural networks. A genetic algorithm (GA) evolves a Lindenmayer System (L-System) that is used to design the neural network’s architecture. This basic neural codification confers scalability and search space reduction in relation to other methods. Furthermore, the system uses a parallel genome scan engine that increases both the implicit parallelism and convergence of the GA. The fitness function of the NEA rewards economical artificial neural networks (ANNs) that are easily implemented. The NEA was tested on five real-world classification datasets and three well-known datasets for time series forecasting (TSF). The results are statistically compared against established stateof- the-art algorithms and various forecasting methods (ADANN, ARIMA, UCM, and Forecast Pro®). In most cases, our NEA outperformed the other methods, delivering the most accurate classification and time series forecasting with the least computational effort. These superior results are attributed to the improved effectiveness and efficiency of NEA in the decisionmaking process. The result is an optimized neural network architecture for solving classification problems and simulating dynamical systems.
Logo do RepositórioLogo do Repositório
Nossas Redes:

DSpace software copyright © 2002-2025 LYRASIS

  • Configurações de Cookies
  • Política de Privacidade
  • Termos de Uso
  • Entre em Contato
Brasão UFPA