NEURAL NETWORKS

2) a) Write the algorithm (in steps)used in Self Organizing Feature Map (SOFM) to update the weights of the Feature map. b) Our aim is to classify the three classes, namely C - 608 000 2 2 C = 1/2 and C, from each other To achieve this goal, we are going to use a self organizing lattice network structure that is represented in two dimensional (in 3x3 matrix) form. Assume that the initial centers are given as follows 0,2 0.8 - 0.1 -0.2 0.4 -0.1 -0.2 0.1 0.1) 0.4 0.6 (0.2 0.5 0.1 0.6 0.3 0.1 0.1 0.4 0 -02 0 0.7 07 and also assume that the learning parameter 1-09 and the neighborhood function A-1 update the weights (using SOFM algorithm) for one iteration using the first element of each class as a training data(as given below) X -1880