Self-organizing neural networks: convergence properties
Roberto Horowitz, Luis Alvarez · 2002
The convergence properties of a class of self-organizing neural networks, introduced and popularized by Kohonen, are analyzed using the ODE approach. It is shown that Kohonen's learning law converges to the best locally affine feature map. A new integrally distributed self-organizing learning law is proposed which converges to the equiprobable feature map for inputs with arbitrary random probability distribution functions.