FORCED FORMATION OF A GEOMETRICAL FEATURE SPACE BY A NEURAL-NETWORK MODEL WITH SUPERVISED LEARNING

To investigate necessary conditions for the object recognition by simu lations using neural network models is one of ways to acquire suggesti ons for understanding the neuronal representation of objects in the br ain. In the present study, we trained a three layered neural network t o form a geometrical feature representation in its output layer using back-propagation algorithm. After training using 73 learning examples, 65 testing patterns made by various combinations of above features co uld be recognized with the network at a rate of 95.3% appropriate resp onse. We could classify four types of hidden layer units on the basis of effects on the output layer.