A two-layer learning method for radial basis function networks using combined genetic and regularised OLS algorithm
A two-layer learning method for radial basis function networks using combined genetic and regularised OLS algorithm
The paper presents a novel two-layer learning method for radial basis function (RBP) networks. At the lower layer, a regularised orthogonal least squares (ROLS) algorithm is employed to construct RBF networks while the two key learning parameters , the regularisation parameter and hidden node's width, needed by the ROLS algorithm are optimized using the genetic algorithm at the higher layer. Networks constructed by this learning method have superior generalisation properties, and the computational complexity of the method is reasonable. Nonlinear time series modelling and prediction is used as an example to demonstrate the effectiveness of this hierarchical learning approach.
245-249
Chen, S.
9310a111-f79a-48b8-98c7-383ca93cbb80
Wu, Y.
84854e37-ada6-4cc8-995f-6ce5ebc77423
Alkadhimi, K.
8d6a3019-7d9c-4e5c-bddd-6fcc0274cc10
1995
Chen, S.
9310a111-f79a-48b8-98c7-383ca93cbb80
Wu, Y.
84854e37-ada6-4cc8-995f-6ce5ebc77423
Alkadhimi, K.
8d6a3019-7d9c-4e5c-bddd-6fcc0274cc10
Chen, S., Wu, Y. and Alkadhimi, K.
(1995)
A two-layer learning method for radial basis function networks using combined genetic and regularised OLS algorithm.
1st IEE/IEEE International Conference on GALESIA, Sheffield, United Kingdom.
12 - 14 Sep 1995.
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
The paper presents a novel two-layer learning method for radial basis function (RBP) networks. At the lower layer, a regularised orthogonal least squares (ROLS) algorithm is employed to construct RBF networks while the two key learning parameters , the regularisation parameter and hidden node's width, needed by the ROLS algorithm are optimized using the genetic algorithm at the higher layer. Networks constructed by this learning method have superior generalisation properties, and the computational complexity of the method is reasonable. Nonlinear time series modelling and prediction is used as an example to demonstrate the effectiveness of this hierarchical learning approach.
Text
c-galesla1995
- Author's Original
Restricted to Repository staff only
Request a copy
More information
Published date: 1995
Additional Information:
1st IEE/IEEE International Conference on GALESIA (Sheffield, UK), Sept. 12-14, 1995. Event Dates: Sept. 12-14, 1995 Organisation: IEE/IEEE
Venue - Dates:
1st IEE/IEEE International Conference on GALESIA, Sheffield, United Kingdom, 1995-09-12 - 1995-09-14
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 251090
URI: http://eprints.soton.ac.uk/id/eprint/251090
PURE UUID: ccd8670e-f489-43f6-bc58-ebd763bb7f4e
Catalogue record
Date deposited: 12 Oct 1999
Last modified: 14 Mar 2024 05:09
Export record
Contributors
Author:
S. Chen
Author:
Y. Wu
Author:
K. Alkadhimi
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics