Exercise 3 [5 points]. Coordinate Descent for Linear Regression. We would like to solve the following linear regression problem M minimize > (y() - wix())2 (2) where we RWx] and () ( RX using coordinate descent. a) [2 points]. In the current iteration, wx is selected for update. Please prove the following update rule: Wk + VK E {1, 2, . .., N} (3) b) [3 points]. Prove that the following update rule for wx is equivalent to Eq. (3). Wk old , Why (4) + wp (5) "(i) + () + (wpd - wx)up) Vic {1, 2, ... M ). (6) where ro is the residual N p(i) = y() (7) Compare the two update rules. Which one is better and why