Repeat problem 10, while choosing as threshold the values of 0.2, 0.4, 0.2, 0.7. What are the…
Repeat problem 10, while choosing as threshold the values of 0.2, 0.4, 0.2, 0.7. What are the main differences from the result obtained in problem 8?
We need to store the fundamental memory given by the vector V = [1 1 −1 1] in a four-node Hopfield network with zero threshold values. Using the different steps of training, deduce the structure of the network, extract the energy levels for each possible state introduced to the network and display the transition table and the transition diagram.
Repeat part (a) of problem 6 with two RBF networks: one with Gaussian MFs and the other with exponential MFs. Compare the outcome of both networks. Compare the performance of the networks with that of problem 6 in terms of execution time and accuracy (mean squared errors).
We need to train an MLP network for obtaining the output of the following two-toone mapping function:
y(x1, x2) = sin(2px1)cos(0.5px2)
(a) Set up two sets of data, each of which consists of 100 input–output patterns, one for network training and the other for testing. The input–output data are obtained by randomly varying the input variables (x1, x2) within the interval [−1, 1] × [−1, 1].
(b) First, fix the number of hidden neurons to two (equal to the number of input nodes) and analyze the performance of the obtained network.
(c) Analyze the performance of the network with more and then with fewer hidden nodes.