neural network
3) The subsets of the classes and C are given as X =- - (310) -02-1) respectively. The aim is to classify the elements of each class. and X = Assuming that n=0.1, the desired value for C, is 1 and the desired value for Cis- 1. a) obtain the synaptic weights of a neuron for the learning set using perceptron algorithm. b) Use the synaptic weights found in part a) to check whether the elements belong to or C Hint: If you cannot reach to the local or global solution till the end of fourth epoch, stop the training algorithm in part a). =3000) (10)-20 4) a) Write the algorithm (in steps)used in Self Organizing Feature Map (SOFM) to update the weights of the Feature map. b) Comment about the selection of the learning parameter (n) and the neighborhood function (A) as the iteration increases. c) Our aim is to classify the three classes C.C, and C. To achieve this goal, we used the self organizing lattice network structure that is represented in two dimensional (in 3x3 matrix) form. After 6000 iterations, the centers and also which class is represented by each center is found and given below. Using these centers, find which class each testing data element of T belongs to. 1.4 -1.3 -1.1 -0.9 1.2 -0.3 1.1 1.7 -1.8 -0.7 -0.5) 1.1 1.4 0.7 (1.6 CC 1.7 C CandT = 1.4 LC, C, 0] (0.4) 1.6 1.3 0.6 -0.2 -1.2 0.1 - 1.1 - 1.0 0.95 -0.45 1.55 1.05 1.0 (0.1) 1.2 (1.5) -0.9 1.2 0.2 1.05 3) The subsets of the classes and C are given as X =- - (310) -02-1) respectively. The aim is to classify the elements of each class. and X = Assuming that n=0.1, the desired value for C, is 1 and the desired value for Cis- 1. a) obtain the synaptic weights of a neuron for the learning set using perceptron algorithm. b) Use the synaptic weights found in part a) to check whether the elements belong to or C Hint: If you cannot reach to the local or global solution till the end of fourth epoch, stop the training algorithm in part a). =3000) (10)-20 4) a) Write the algorithm (in steps)used in Self Organizing Feature Map (SOFM) to update the weights of the Feature map. b) Comment about the selection of the learning parameter (n) and the neighborhood function (A) as the iteration increases. c) Our aim is to classify the three classes C.C, and C. To achieve this goal, we used the self organizing lattice network structure that is represented in two dimensional (in 3x3 matrix) form. After 6000 iterations, the centers and also which class is represented by each center is found and given below. Using these centers, find which class each testing data element of T belongs to. 1.4 -1.3 -1.1 -0.9 1.2 -0.3 1.1 1.7 -1.8 -0.7 -0.5) 1.1 1.4 0.7 (1.6 CC 1.7 C CandT = 1.4 LC, C, 0] (0.4) 1.6 1.3 0.6 -0.2 -1.2 0.1 - 1.1 - 1.0 0.95 -0.45 1.55 1.05 1.0 (0.1) 1.2 (1.5) -0.9 1.2 0.2 1.05
