Pruning back propagation neural networks using modern stochastic optimization techniques
Pruning back propagation neural networks using modern stochastic optimization techniques
Approaches combining genetic algorithms and neural networks have received a great deal of attention in recent years. As a result, much work has been reported in two major areas of neural network design: training and topology optimisation. This paper focuses on the key issues associated with the problem of pruning a multilayer perceptron using genetic algorithms and simulated annealing. The study presented considers a number of aspects associated with network training that may alter the behaviour of a stochastic topology optimiser. Enhancements are discussed that can improve topology searches. Simulation results for the two mentioned stochastic optimisation methods applied to non-linear system identification are presented and compared with a simple random search.
76-98
Stepniewski, Slawomir W.
ce753ba1-2869-45af-b0f4-213e0b68ad94
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
1997
Stepniewski, Slawomir W.
ce753ba1-2869-45af-b0f4-213e0b68ad94
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
Stepniewski, Slawomir W. and Keane, Andy J.
(1997)
Pruning back propagation neural networks using modern stochastic optimization techniques.
Neural Computing and Applications, 5 (2), .
(doi:10.1007/BF01501173).
Abstract
Approaches combining genetic algorithms and neural networks have received a great deal of attention in recent years. As a result, much work has been reported in two major areas of neural network design: training and topology optimisation. This paper focuses on the key issues associated with the problem of pruning a multilayer perceptron using genetic algorithms and simulated annealing. The study presented considers a number of aspects associated with network training that may alter the behaviour of a stochastic topology optimiser. Enhancements are discussed that can improve topology searches. Simulation results for the two mentioned stochastic optimisation methods applied to non-linear system identification are presented and compared with a simple random search.
Text
step_97a.pdf
- Accepted Manuscript
More information
Published date: 1997
Identifiers
Local EPrints ID: 21081
URI: http://eprints.soton.ac.uk/id/eprint/21081
PURE UUID: 30d812ea-6ace-4572-b801-f29af456c72e
Catalogue record
Date deposited: 31 Oct 2006
Last modified: 16 Mar 2024 02:53
Export record
Altmetrics
Contributors
Author:
Slawomir W. Stepniewski
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
