a) A Machine Learning practitioner needs to estimate the parameters 0 of a model, given some data X. Suppose, that the parameters can take only two values, 01 or 02 and that it is given that p(X191) = 0.8, p(X102) = 0.5, p(01) = 0.4, and p(92) = 0.6. i) What does each of the terms p(X10i) and e 11,2) denote? ii) Is it consistent with the laws of probabilities that p(X101) + p(X102) = 0.8 + 0.5 > 1? Is it required that p(911X) + p(021X) = 1? Explain briefly why. iii) Which model will he/she choose under the Maximum Likelihood criterion? Which model will he/she choose under the Maximum Aposteriori Probability criterion? Show your calculations.