Computational Optimization

Due: September 7

Problem 1
Given the problem
min f(x) = (x1-3)2 + (x2 - 4)2
s.t. 2x1 + x2 - 6 < 0
x1 - x2 - 4 < 0
x1 ≥ 1.8
x2 ≥ 0

Plot the contours for f = 0, 1, 2, 4, 6, 8, 10, and the feasible region. From inspection, what is the
optimal solution?

Problem 2
Without plotting the feasible region, determine whether the following inequalities define a
convex feasible region
x1 + 2 x2 ≤ 3
− exp ( 2 x1 ) + x2 ≥ 0
0 ≤ x1 ≤ 3

Problem 3
Consider the n-particle cluster of molecules interacting through the classical Lennard-Jones
potential:
n −1 n n −1 n
1 2
f = ∑ ∑ f ij = ∑ ∑ 12 − 6
i =1 j =i+1 i=1 j=i +1 rij rij
where rij > 0 is the 3-dimensional Euclidean distance between particles i and j.

Prove the following properties:

1
Property 1: The repulsive term 12 is a convex function of rij .
rij
 2
Property 2: The attractive term − 6 is a concave function of rij .
rij
* *
Property 3: The minimizer of f ij with respect to rij occurs at rij = 1 with a value of f ij = −1 .
Property 4: f ij is a convex function of rij for 0 < rij ≤ rˆ and concave for rˆ ≤ rij , where
1
 13 6
rˆ =   .
7

Problem 4
Solve the following problems by Newton’s method considering at least 3 iterations.

min f ( x ) =1 + x1 + x2 + x3 + x4 + x1 x2 + x1 x3 + x1 x4
(a)
+ x2 x3 + x2 x4 + x3 x4 + ( x1 ) + ( x2 ) + ( x3 ) + ( x4 )
2 2 2 2

Starting from x 0 =[ −3 −30 −4 −0.1] T

(b) min f ( x ) x1 ( x2 ) ( x3 ) ( x4 ) exp ( − x1 − x2 − x3 − x4 ) 

=
2 3 4

Starting from x 0 = [ 3 4 0.5 1]

T

Problem 5
Given the data set of 25 samples on weight, age and blood fat content, please propose a linear
model for predicting the blood fat content (Y) by relating it to age (X1) and weight (X2), i.e. Y =
𝜃𝜃0 + 𝜃𝜃1 * X1 + 𝜃𝜃2 * X2.
a) Formulate an unconstrained optimization problem to minimize the mean square error
(MSE), and solve the problem using Newton’s method starting from θ0 = [ 0 0 0] to
T

determine the parameters of the linear regression model.

b) Verify your solution by solving it with the analytical method. You might use MATLAB,
Mathematica, or any tool that you are familiar with for matrix operations.
Data Set:
Weight (kilograms) Age (Years) Blood fat content
X1 X2 Y
84 46 354
73 20 190
65 52 405
70 30 263
76 57 451
69 25 302
63 28 288
72 36 385
79 57 402
75 44 365
27 24 209
89 31 290
65 52 346
57 23 254
59 60 395
69 48 434
60 34 220
79 51 374
75 50 308
82 34 220
59 46 311
67 23 181
85 37 274
55 40 303
63 30 244