1. We are conducting a study to predict a Boolean variable y from two features 931 and :32 that only take integer values, 0 g 931 g 4 and 0 3 3:2 3 3. As our hypothesis class 'H we take the class characteristic functions of closed rectangles with integer vertices contained in X = {(x, y) 6 [0,4] X [0, 3], ac and y are integers} (including degen- erate ones7 like [0,2] X {2}). Given the training set a) What is the empirical risk of the rectangle [0, 2] x [0,2]? In) Find some Empirical Risk Minimizer (there are many). c) Does the realizability assumption hold? d) What is the true risk of the ERM found in part (b) if the actual distribution of the features is uniform over X and Web are f (3:,y) = 1 if a: is even and f (:17, y) = 0 if m is odd? A