Let us consider the following table representing (up to normalization constant) a joint distribution p for (X, Y). Notice that Y can take on three different values. x=b 0.1 X=a X=c y=1 0.5 y=2 0.1 y=0 0.55 0.1 0.4 0.55 0.4 0.6 Let l: {0, 1, 2} x {0, 1, 2} R be the loss function defined as: (y, y) = 0 Vy, (0, 2) = (2,0) = 1/2, (0, 1) = 1(1,0) = 1(1,2) = 1(2, 1) = 1. Find a classifierf that minimizes the quantity Ex,Y)~p ((S(X), Y)). Prove that your classifier is indeed a minimizer.