|
Descripción
|
|
|---|---|
| This article presents an evolutionary algorithm to autonomously construct full-connected multilayered feedforward neural architectures. This algorithm employs grammar-guided genetic programming with a context-free grammar that has been specifically designed to satisfy three important restrictions. First, the sentences that belong to the language produced bythe grammar only encode all valid neural architectures. Second, full-connected feedforwardneural architectures of any size can be generated. Third, smaller-sized neural architecturesare favored to avoid overfitting. The proposed evolutionary neural architectures construction system is applied to compute the terms of the two sequences that define the three-term recurrence relation associated with a sequence of orthogonal polynomials. This application imposes an important constraint: training datasets are always very small. Therefore, an adequate sized neural architecture has to be evolved to achieve satisfactory results, which arepresented in terms of accuracy and size of the evolved neural architectures, and convergencespeed of the evolutionary process. | |
|
Internacional
|
Si |
|
JCR del ISI
|
Si |
|
Título de la revista
|
Annals of Mathematics And Artificial Intelligence |
|
ISSN
|
1012-2443 |
|
Factor de impacto JCR
|
0,807 |
|
Información de impacto
|
Datos JCR del año 2016 |
|
Volumen
|
|
|
DOI
|
10.1007/s10472-018-9601-2 |
|
Número de revista
|
|
|
Desde la página
|
1 |
|
Hasta la página
|
24 |
|
Mes
|
SEPTIEMBRE |
|
Ranking
|
|