Title :
On the efficiency of the orthogonal least squares training method for radial basis function networks
Author :
Sherstinsky, Alex ; Picard, Rosalind W.
Author_Institution :
Media Lab., MIT, Cambridge, MA, USA
fDate :
1/1/1996 12:00:00 AM
Abstract :
The efficiency of the orthogonal least squares (OLS) method for training approximation networks is examined using the criterion of energy compaction. We show that the selection of basis vectors produced by the procedure is not the most compact when the approximation is performed using a nonorthogonal basis. Hence, the algorithm does not produce the smallest possible networks for a given approximation error. Specific examples are given using the Gaussian radial basis functions type of approximation networks
Keywords :
feedforward neural nets; interpolation; learning (artificial intelligence); least squares approximations; Gaussian radial basis functions type; approximation networks; energy compaction; interpolation; nonorthogonal basis; orthogonal least squares training; radial basis function networks; Approximation algorithms; Automatic control; Compaction; Fuzzy logic; Image coding; Interpolation; Least squares approximation; Least squares methods; Radial basis function networks; Vectors;
Journal_Title :
Neural Networks, IEEE Transactions on
