3. The following image illustrates the topology of a simple feedforward neural network that has a single sensing neuron (Neuron 1), a single hidden processing neuron (Neuron 2), and a single processing output neuron (Neuron 3).
w2,0=0.1
w3,0=0.1
w2,1=0.2
w3,2=0.3
)
a. Assuming that the processing neurons use logistic activation functions, that the input to the network is Neuron 1 = 0.2 and that the desired output for this input is 0.7: i. Calculate the output generated by the network in response to this input ii. Calculate the S values for each of the neurons in the network (i.e.,83, 82) iii. Using the values you calculated above, calculate the sensitivity of the error of the network to changes in each of the weights of the network: i.e. 08/0W3,2, 08/0W3,0,08/0W2,1,08/0W2,0
iv.
Assuming a learning rate of a = 0. 1, calculate the updated values for
2,0
processing of this single training example Calculate the reduction in the error of the network for this example using the new weights, compared with using the original weights.
V.
