Question: 3 Least squares with a linear function of the parameters w is called Linear Regression. You are given N observed examples (x1,yl),(x2,y2),,(xN,yN), and you decide

3 Least squares with a linear function of the parameters w is called "Linear Regression". You are given N observed examples (x1,yl),(x2,y2),,(xN,yN), and you decide to use Linear Regression to construct a model as y=WT. Which of the following statement is correct? (10 points) (a). You minimize the following optimization problem to find the best w : minwi=1N(yixiTw)2 (b). You minimize the following optimization problem to find the best w: minwi=1N(yixiTw)2+w2. (c). You can use a gradient descent algorithm to solve the related optimization problem iteratively to obtain the values of w. (d). You can compute w using a closed-form solution w=(XTX)XTy where y is a column vector containing all observed yi 's, and X is a matrix where the i-th row represents xi as a row vector
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
