BDBComp
Parceria:
SBC
Usando Redes Neurais para Acelerar o Processo de Otimização de Funções com Alto Custo Computacional

Hilton Cleber PietrobomKarl Heinz Kienitz

In many classic optimization problems large computational effort is needed to determine a solution. An alternative is use surrogates (simplified models and / or functions). This contribution presents a neural-network-based tool that generates a sequence of real objective function approximations and manages the use of these approximations. Numerical results are given which show that such tool may reduce efficiently the computational effort to determine the solution, without sacrificing convergence. Clique no link abaixo para buscar o texto completo deste trabalho na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato: bdbcomp@lbd.dcc.ufmg.br
     Mantida por:
LBD