Please show your work. 1. A multi-layer feedforward ANN is shown below. Using a training record, initial weights and biases that are given in the table below, train the ANN only for the first iteration by taking the following steps: 1) Activate the ANN forward calculate the net input and output values for the hidden neurons (1 & 2) and then the output neuron (3). Transfer functions are sigmoid in the hidden and linear in the output layer 2) Back-propagation - calculate weight changes for the hidden-output, and then input-hidden. Set learning rate to 0.9 3) Update all the weights and biases with the values you obtained in 2) 4) Activate the ANN again with the updated weights and biases and see how much the error changes 31 2 X1 X2 x3 target W11 w21 W12 W22 W13 W23 W31 32 b b2 b 1 0.2 -0.3 0.4 0.1 -0.5 0.2 -0.3 -0.2 0.4 -0.2 -0.1