Neural network architecture selection: new Bayesian perspectives in predictive modelling - Application to a soil hydrology problem

Authors
Vila, JP Wagner, V Neveu, P Voltz, M Lagacherie, P
Citation
Jp. Vila et al., Neural network architecture selection: new Bayesian perspectives in predictive modelling - Application to a soil hydrology problem, ECOL MODEL, 120(2-3), 1999, pp. 119-130
Citations number
29
Categorie Soggetti
Environment/Ecology
Journal title
ECOLOGICAL MODELLING
ISSN journal
03043800 → ACNP
Volume
120
Issue
2-3
Year of publication
1999
Pages
119 - 130
Database
ISI
SICI code
0304-3800(19990817)120:2-3<119:NNASNB>2.0.ZU;2-#
Abstract
The aim of this paper is to present to the community of ecologists concerne d with predictive modelling by feedforward neural network, a new statistica l approach to select the best neural network architecture (number of layers , number of neurons per layer and connectivity) in a set of several candida te networks. The interest of this approach is demonstrated on a soil hydrol ogy problem. (C) 1999 Elsevier Science B.V. All rights reserved.