(20 points) Consider training a perceptron using the datapoints in the table below, presented in this order. Instance 1 2 3 4 5 6 7 8 Label y +1 -1 +1 -1 +1 -1 +1 +1 Data (x1, x2) 10,10 0,0 8,4 3,3 4,8 0.5, 0.5 4,3 2,5 Given an initial set of weights w = (1, 1) and bias b = 0, show each step of the perceptron algorithm for the above sequence of instances over one epoch. This includes computation of the activation and adjustment of the weights after each instance. What is the accuracy of the perceptron after this first epoch? Will this perceptron eventually converge on a model with zero error for the training data? Why or why not