A global-local artificial neural network with application to wave overtopping prediction

Wedge, David C. and Ingram, David M. and McLean, David A. and Mingham, Clive G. and Bandar, Zuhair A. (2005) A global-local artificial neural network with application to wave overtopping prediction. In: UNSPECIFIED UNSPECIFIED.

Full text not available from this repository.

Abstract

We present a hybrid Radial Basis Function (RBF) - sigmoid neural network with a three-step training algorithm that utilises both global search and gradient descent training. We test the effectiveness of our method using four synthetic datasets and demonstrate its use in wave overtopping prediction. It is shown that the hybrid architecture is often superior to architectures containing neurons of a single type in several ways: lower errors are often achievable using fewer hidden neurons and with less need for regularisation. Our Global-Local Artificial Neural Network (GL-ANN) is also seen to compare favourably with both Perceptron Radial Basis Net (PRBFN) and Regression Tree RBFs

Item Type: Book Section
Additional Information: Citation: Wedge, D.C. et al. A global-local artificial neural network with application to wave overtopping prediction. In Artificial Neural Networks: Formal Models and Their Applications - ICANN 2005, Springer, 2005, pp. 109-114.
Divisions: Faculties > Faculty of Science and Engineering > Department of Computing, Mathematics & Digital Technology
Faculties > Faculty of Science and Engineering > Department of Computing and Mathematics: Centre for Mathematical Modelling and Flow Analysis (CMMFA)
Legacy Research Institutes > Dalton Research Institute > Environmental Science
Date Deposited: 14 Jul 2008 15:11
Last Modified: 20 Jul 2016 01:15
URI: http://e-space.mmu.ac.uk/id/eprint/31954

Actions (login required)

Edit Item Edit Item