Check image for Question.

2 Consider the squared loss L(X,w,y)=21Xwy2 for data matrix XRND, weights wRD1, and outputs yRN1. (a) Find the expression for gradient wL(X,w,y) and minimizer of this loss, argminwL(X,w,y). (Hint: See the example on page 96 of Goodfellow I, Bengio Y, Courville A. Deep learning. MIT press, Link .) (b) Take w0 as the initialization for gradient descent with step size and show an expression for the first and second iterates w1 and w2 only in terms of ,w0,X,y. (c) Generalize this to show an expression for wk in terms of ,w0,X,y,k. (d) Write a pseudo code for calculating the wk in terms of ,w0,X,y,k