Improving the Generalization Properties of Radial Basis Function Neural Networks
Keyword(s):
An important feature of radial basis function neural networks is the existence of a fast, linear learning algorithm in a network capable of representing complex nonlinear mappings. Satisfactory generalization in these networks requires that the network mapping be sufficiently smooth. We show that a modification to the error functional allows smoothing to be introduced explicitly without significantly affecting the speed of training. A simple example is used to demonstrate the resulting improvement in the generalization properties of the network.
2013 ◽
Vol 4
(1)
◽
pp. 56-80
◽
2004 ◽
Vol 02
(03)
◽
pp. 511-531
◽
2008 ◽
Vol 205
(2)
◽
pp. 908-915
◽
Keyword(s):
2007 ◽
Vol 31
(7)
◽
pp. 1271-1281
◽
2013 ◽
Vol 221
◽
pp. 503-513
◽
Keyword(s):
Keyword(s):