i) Derive the training algorithm for Madaline below training pattern Question 1 (Total of 20 marks) A simple MADALINE network is illustrated in Figure 1. The outputs of the two hidden ALINES, Z1 and 22, are determined by signals from the same input units X1 and As with the Adalines, each output signal is the result of applying a threshold function to the unit's net input ii) Apply the above training algorithm to solve XoR problem for and determine all the weights and the biases: + 1 1 -1 1 - 1 1 1 -1 iii) The positive response region of the above Madaline algorithm is the union of the regions where each of hidden units have positive responses. Determine the boundary lines from hid- den units 21, 22, and plot the positive response regions in a graph for Madaline. W12 w Figure 1: Madaline network The activation function for units 21, 22, and Y is: f(x) = 1 - 1 if x = 0; if