Please explain as best you can how to get the answer(with theory) and I will give you a thumbs up. Don't simply use chat GPT.

(b) Fig. Q2 shows a neural network with two input nodes, two hidden nodes, and one output node. The inputs with value 1.0 to the hidden nodes implement the bias terms. All processing nodes use sigmoid nonlinearity. Fig. Q2 (i) Determine the maximum number of classes the network can classify. (2 marks) (ii) Determine the maximum number of decision boundaries the network can produce. (3 marks) (iii) Explain why the bias terms are essential if the network is used for classification