Starting from:

$30

COMP6211E-Homework 1 Solved

                  Optimization for Machine Learning   


Theoretical Problems 

1.    Let             4, where x ∈Rd. Find its conjugate f∗(x).

2.    Let x ∈R and y ∈R+. Is f(x,y) = x2/y a convex function? Prove your claim.

3.     Consider the convex set C = {x ∈Rd : kxk∞ ≤ 1}. Given y ∈Rd, compute the projection projC(y).

4.    Compute ∂f(x) for the following functions of x ∈Rd

•   f(x) = kxk2

•   f(x) = 1(kxk∞ ≤ 1)

•   f(x) = kxk2 + kxk∞

5.     Consider the square root Lasso method. Given X ∈ Rn×d and y ∈ Rn, we want to find w ∈Rd to solve

                                                            ,                                                              (1)

                                                                          subject to ξj ≥ wj,           ξj ≥−wj                  (j = 1,...,d).                                            (2)

Lasso produces sparse solutions. Define the support of the solution as

S = {j : w∗,j 6= 0}.

Write down the KKT conditions under the assumption that Xw∗ 6= y. Simplify in terms of S,XS,XS¯,y,wS. Here XS contains the columns of X in S, XS¯ contains the columns of X not in S, and wS contains the nonzero components of w∗.

Programming Problem (4 points)

We consider ridge regression problem with randomly generated data. The goal is to implement gradient descent and experiment with different strong-convexity settings and different learning rates.

•    Use the python template “prog-template.py”, and implement functions marked with ’# implement’.

•    Submit your code and outputs. Compare to the theoretical convergence rates in class, and discuss your experimental results.

1

More products