Initialising self-organising maps

Abstract

We review a technique for creating Self-Organising Maps (SOMs) in a Feature space which is nonlinearly related to the original data space. We show that convergence is remarkably fast for this method. By considering the linear feature space, we show that it is the interaction between the overcomplete basis in which learning takes place and the mixture of one-shot and incremental learning which comprises the method that gives the method its power. We illustrate the method on real and artificial data sets. © Springer-Verlag 2003.