a) What is the meaning of Perceptron? Plot a sample of it, and discuss why it could solve some problems like AND and OR logics, while it could not solve the problem of XOR logic? b) List three different possible activation functions used in the artificial neural network. c) Generally, the error of any system is calculated as the difference between the desired output and the actual output. As we mentioned in our lectures that there is no obvious way to know what is the desired output of the hidden layer should be. Therefore, how the error will be taken?!! d) Plot an artificial neural network with 2 inputs (x, x), 2 outputs (ya, yb), and 3 neurons single hidden layer. Place all the required links, biases, and weights.