Machine Learning

3. The instance set ,1)1,1), (-1,0), (1,-2) R2 with label set , 1,-1,- is linearly separable, and this can be done with bias b = O. Consider the Perceptron algorithm applied to (X,). (a) Find the minimum B-minW Vi, yW X, 2 1) when you restrict to the case that the bias b-0. Determine an upper bound on the number of iterations necessary for the Perceptron algorithm on this (X,)). (b) Use the Perceptron algorithm to find a hyperplane +b(here is the final hat separates (x, ). Keep track of how many iterations were necessary: the number T