A Study of Parallel Self-Organizing Map

A new Self-Organizing Map is proposed for information parallel processing purpose. In this model, Parallel-SOM, there are two separate layers of neurons connected together. The number of neurons in both layer and connections between them is equal to the number of total elements of input signals. The weight updating is managed through a sequence of operations among some unitary transformation and operation matrixes. So the conventional repeated learning procedure is modified to learn just once. This research presents an algorithm developed to realize this new learning method. With a typical classification example, the performance of Parallel-SOM demonstrated convergence results similar to Kohonen's model. Theoretic analysis and proofs also show some interesting properties of Parallel-SOM. As the paper pointed out, in classical computation sense, the contribution of such a network may not be so significant. But its parallel mode may be interesting for quantum computation