Question: Consider the following four-class problem which contains 8 tuples, each having two inputs p and the target class t. You are required to train a perceptron network to solve this problem using the Hebbian Learning Rule given in Eq.1. You are required to train the network for the first 3 iterations.
To start training, initial weights and biases are given as following:
W(0) = [-8, p=[]= ==B, -]
Instructions: Use the Hebbian Learning weights update mechanism provided as following: w = w + nxy (1), where x and y are the ith input and corresponding target output, respectively. Network weights are denoted by w for the ith input and the layer, whereas the learning rate is defined as n = 1.
Perform calculations for the first 5 iterations.
Hint: Problem P4.5. Page 4-30. Neural Network Design 2nd Edition