= 2. (15 points) Given two normal distributions with proportions a = 1 (2,5)' and 2 = (4, 1)', and common covariance matrix 1/5 and a2 = 4/5, means = x-(). 3 1 1 2 determine the linear decision boundary for the two Bayes classification regions. Express your answer in the form y = mx + b. How will the decision boundary change if we modify this so that a = a2 = 1/2? Explain.