13.13. Self-Organizing Feature Maps
SOFMs learn to recognize groups of similar input vectors in such a way that neurons physically near each other in the layer respond to similar input vectors. These networks learn to classify input vectors according to how they are grouped in the input space.
SOFMs make use of competitive transfer function that accepts a net input vector and returns neuron outputs of 0 for all neurons except for the winner whose net input is maximum. The winner’s output is 1. They differ from competitive layers in that learning function updates neighbouring neurons also based on their distance from the winning neuron. Thus, SOFMs learn both the distribution and topology of the input vectors they are trained on.
The neurons ...
Get MATLAB® and Its Applications in Engineering: [Based on MATLAB 7.5 (R2007b)] now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
