×

A sequential learning scheme for function approximation using minimal radial basis function neural networks. (English) Zbl 1067.68586

Summary: This article presents a sequential learning algorithm for function approximation and time-series prediction using a minimal radial basis function neural network (RBFNN). The algorithm combines the growth criterion of the resource-allocating network (RAN) of Platt (1991) with a pruning strategy based on the relative contribution of each hidden unit to the overall network output. The resulting network leads toward a minimal topology for the RBFNN. The performance of the algorithm is compared with RAN and the enhanced RAN algorithm of Kadirkamanathan and Niranjan (1993) for the following benchmark problems: (1) hearta from the benchmark problems database PROBEN1, (2) Hermite polynomial, and (3) Mackey-Glass chaotic time series. For these problems, the proposed algorithm is shown to realize RBFNNs with far fewer hidden neurons with better or same accuracy.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1006/dspr.1994.1016 · doi:10.1006/dspr.1994.1016
[2] DOI: 10.1049/ip-d.1993.0058 · Zbl 0799.68161 · doi:10.1049/ip-d.1993.0058
[3] DOI: 10.1162/neco.1993.5.6.954 · doi:10.1162/neco.1993.5.6.954
[4] DOI: 10.1162/neco.1989.1.2.281 · doi:10.1162/neco.1989.1.2.281
[5] DOI: 10.1016/S0893-6080(05)80038-3 · doi:10.1016/S0893-6080(05)80038-3
[6] DOI: 10.1162/neco.1991.3.2.213 · doi:10.1162/neco.1991.3.2.213
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.