The parity problem returns 1 if the number of inputs (each input is either 1 or 0) that are 1 is even, and 0 otherwise. Can a Perceptron learn this problem for 3 inputs? If so, devise such a network. If not, explain why