Neural Networks Elitist Evolution

Título:

Neural Networks Elitist Evolution

Autor:

Vinuesa, Hernán

Colaboradores:

; Lanzarini, Laura Cristina

Temas:

ROBÓTICA

En:

International Conference on Information Technology Interfaces (29th : 2007 : Duvrovnik)

Resumen:

This paper presents an elitist evolving strategy, which allows obtaining a controller, based on a neural network capable of commanding an autonomous robot. In order to reduce the detrimental crossover effect, we propose to use a strategy to create several children for each parent pair, selecting properly the way of making the replacement. The results obtained that, though the number ofchildren is high, the quantity of fitness tests carried out is actually lower than that of aconventional evolving algorithm. In this way, we propose an alternative that reduces thecomputational cost of the process, reaching at a suitable response for the problem resolution.

URL/DOI:

Medio:

Soporte electrónico

Tipo de documento:

Artículo

Descripción física:

Datos electrónicos (1 archivo: 208 KB)

Idioma:

Español

Publicación:

, 2007

Puede solicitar más fácilmente el ejemplar con: A0228

Ver estantes

Título:

Neural Networks Elitist Evolution

Autor:

Vinuesa, Hernán

Colaboradores:

; Lanzarini, Laura Cristina

Temas:

ROBÓTICA

En:

International Conference on Information Technology Interfaces (29th : 2007 : Duvrovnik)

Resumen:

This paper presents an elitist evolving strategy, which allows obtaining a controller, based on a neural network capable of commanding an autonomous robot. In order to reduce the detrimental crossover effect, we propose to use a strategy to create several children for each parent pair, selecting properly the way of making the replacement. The results obtained that, though the number ofchildren is high, the quantity of fitness tests carried out is actually lower than that of aconventional evolving algorithm. In this way, we propose an alternative that reduces thecomputational cost of the process, reaching at a suitable response for the problem resolution.

URL/DOI:

Medio:

Impreso en papel

Tipo de documento:

Artículo

Idioma:

Español

Puede solicitar más fácilmente el ejemplar con: