Exercises Train a 3-input, 1-output artificial neuron to classify apples and oranges in a supervised scheme based on each of the three learning rules. Use five pairs of input training pattens, s and their coresponding target values , for e.g. (s, t) - 1. --1-1). Make a random initial assignment of the weights. Gannon University ECE 474/GECE 574 ANN Fall 2017 Dr. Ram Sundaram 2. After the training process is completed, test your neuron with arbitrary input patterns representing either or neither fruit. Report on the accuracy of classification in the case of each leaming rule. 3. Develop an artificial neural network which extends this simple classification problem to the case when more varieties of fruit must be handled (e.g. apples, oranges, bananas, mangoes, peaches, plums)