Question 1 Answer the following questions:

1. The input to a single-input neuron is 2.0, its weight is 2.3 and its bias is -3. What is the output of the neuron if it has the following transfer functions?

   (1) Hard limit. (2) Linear. (3) Sigmoid f (x) = 1/(1 + exp(x))

2. What would happen if the transfer functions (at the hidden and output layer) in a multi-layer perceptron would be omitted; i.e. if the activation would simply be the weighted sum of all input signals? Explain why this (simpler) activation scheme is not normally used in MLPs although it would simplify and accelerate the calculations for the back-propogation algorithm.

Question 1 (5+5+5+5+5 = 25 marks) Answer the following questions: 1. What is a decision boundary in a pattern classification problem with two input variables? [5 marks] 2. What is generalization of an artificial neural network model? What are the main factors that determine a network's generalization capability? [5 marks] 3. When are the artificial neural networks a good choice for problem solving? [5 marks] 4. The input to a single-input neuron is 2.0, its weight is 2.3 and its bias is -3. What is the output of the neuron if it has the following transfer functions? (1) Hard limit. (2) Linear. (3) Sigmoid f(x) = 1/(1+ exp(-x)) [5 marks] 5. What would happen if the transfer functions (at the hidden and output layer) in a multi-layer perceptron would be omitted; i.e. if the activation would simply be the weighted sum of all input signals? Explain why this (simpler) activation scheme is not normally used in MLPS although it would simplify and accelerate the calculations for the back-propogation algorithm. [5 marks]