Growing radial basis neural networks: merging supervised and unsupervised learning with network growth techniques
- 1 November 1997
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks
- Vol. 8 (6), 1492-1506
- https://doi.org/10.1109/72.641471
Abstract
This paper proposes a framework for constructing and training radial basis function (RBF) neural networks. The proposed growing radial basis function (GRBF) network begins with a small number of prototypes, which determine the locations of radial basis functions. In the process of training, the GRBF network gross by splitting one of the prototypes at each growing cycle. Two splitting criteria are proposed to determine which prototype to split in each growing cycle. The proposed hybrid learning scheme provides a framework for incorporating existing algorithms in the training of GRBF networks. These include unsupervised algorithms for clustering and learning vector quantization, as well as learning algorithms for training single-layer linear neural networks. A supervised learning scheme based on the minimization of the localized class-conditional variance is also proposed and tested. GRBF neural networks are evaluated and tested on a variety of data sets with very satisfactory results.Keywords
This publication has 18 references indexed in Scilit:
- Gradient descent learning of radial basis neural networksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A methodology for constructing fuzzy algorithms for learning vector quantizationIEEE Transactions on Neural Networks, 1997
- Fuzzy algorithms for learning vector quantizationIEEE Transactions on Neural Networks, 1996
- ALADIN: algorithms for Learning and Architecture DetermINationIEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, 1994
- Efficient learning algorithms for neural networks (ELEANNE)IEEE Transactions on Systems, Man, and Cybernetics, 1993
- On the training of radial basis function classifiersNeural Networks, 1992
- Pattern classification using neural networksIEEE Communications Magazine, 1989
- Fast Learning in Networks of Locally-Tuned Processing UnitsNeural Computation, 1989
- Interpolation of scattered data: Distance matrices and conditionally positive definite functionsConstructive Approximation, 1986
- Pattern Recognition with Fuzzy Objective Function AlgorithmsPublished by Springer Science and Business Media LLC ,1981