Unsupervised competitive Kohonen learning

The aim of Kohonen learning is to map similar signals to similar neuron positions. The learning procedure is unsupervised competitive learning where in each cycle the neuron c is to be found with the output most similar to the input signal ... 

Neural networks for unsupervised learning are based on a competitive layer of weights arranged linearly or in a plane (Figure 8.13). If arranged in a plane, the nets are termed a Kohonen network. The peculiarity of this network type is the maintenance of topology or, more general, the pattern of the data vector to be learned. 

---