Introduction to Neural Networks for Java, Session 12

Session Material

A self-organizing map (SOM) has a somewhat different structure than the feedforward networks we've examined so far. A typical SOM is shown here.

The SOM has no hidden layers. A normalization of the input data is performed before passing the data on to the input neurons. Some consider this normalization an additional layer. The output is whichever neuron has the highest activation. The SOM uses a winner take all mentality.

Videos for this Session

Video	Title
	Introduction to Neural Networks for Java(Class 12/16, Part 1/5)
	Introduction to Neural Networks for Java(Class 12/16, Part 2/5)
	Introduction to Neural Networks for Java(Class 12/16, Part 3/5)
	Introduction to Neural Networks for Java(Class 12/16, Part 4/5)
	Introduction to Neural Networks for Java(Class 12/16, Part 5/5)