Cluster learning algorithm of synergetic neural network
Dong Huoming · Journal of Hefei University of Technology · 2002
Synergetic neural network is a new kind of neural network. According to the theory of competitive neural network, it can be divided into two parts: the matching sub-network and the competitive sub-network. And the learning of the matching sub-network is the basic problem of synergetic neural network. There are two ways to enhance the learning efficiency of the matching sub-network: by improving the method of obtaining the adjoining vectors and by improving the method of selecting the prototype vectors. In this paper,the two solutions are discussed, and the application of cluster algorithm in selecting the prototype vectors is studied in detail. Examples prove that the presented two ideas for selecting the prototype vectors are effective in dealing with the images of a set of traffic signs.