Scribd
Upload
40 views2 pages

Homework #2 Solutions

The document summarizes the solutions to homework problems about convex sets and linear programming. It proves that a hyperplane, the intersection of two convex sets, and the intersection of any finite number of convex sets are convex. It also proves that the set of vectors satisfying a system of linear equations Ax=b is convex, and finds the extreme points and optimal solution for a linear program.

Uploaded by

Rajeshwar Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
40 views2 pages

Homework #2 Solutions

The document summarizes the solutions to homework problems about convex sets and linear programming. It proves that a hyperplane, the intersection of two convex sets, and the intersection of any finite number of convex sets are convex. It also proves that the set of vectors satisfying a system of linear equations Ax=b is convex, and finds the extreme points and optimal solution for a linear program.

Uploaded by

Rajeshwar Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Math 354 Summer 2004

Homework #2 Solutions
1 Sketch the set of feasible solutions to the following set of inequalities

−x + y ≤ 2,
2x + y ≤ 2,
x ≥ 0,
y ≤ 1.

Answer: You all got it right, and besides, graphs are hard to draw.

2 Prove that a hyperplane (defined on page 72, a hyperplane is a set of the form {x : aT x = b}
for some vector a and real number b) is a convex set (defined on page 79).
Proof: Let H be a hyperplane. We know that H is of the form {x : aT x = b} for some a and
b. We want to show that H is convex, i.e., for any two points x1 and x2 in H and any point x
on the line segment between x1 and x2 , we want to show that x also lies in H. Now take two
points x1 and x2 in H and let x lie on the line segment between x1 and x2 . We know that
x = λx1 + (1 − λ)x2 for some λ between 0 and 1. Now we just check the hyperplane condition
for x:

aT x = aT (λx1 + (1 − λ)x2 ) ,
= λaT x1 + (1 − λ)aT x2 ,
= λb + (1 − λ)b,
= b,

and this shows that x does indeed lie on the hyperplane, proving the claim.

3 Prove that the intersection of two convex sets is a convex set.


Proof: Let A and B be convex sets. We want to show that A ∩ B is also convex. Take
x1 , x2 ∈ A ∩ B, and let x lie on the line segment between these two points. Then x ∈ A
because A is convex, and similarly, x ∈ B because B is convex. Therefore x ∈ A ∩ B, as
desired.

4 Prove that the intersection of any finite number of convex sets is a convex set. Hint: this
should follow from the previous problem; for example, to show that A1 ∩ A2 ∩ A3 is convex,
write it as (A1 ∩ A2 ) ∩ A3 .
Proof: Let A1 , A2 , . . . , An be convex sets. Then

A1 ∩ A2 ∩ . . . ∩ An = (. . . ((A1 ∩ A2 ) ∩ A3 ) . . .) ∩ An ).

By the previous problem, A1 ∩ A2 is convex, so ((A1 ∩ A2 ) ∩ A3 ) is also convex, etc.


To write up a more formal proof, use induction.
Math 354 Summer 2004
5 Let A be an m × n matrix and let b be a vector in Rm . Prove that the set of vectors x in Rn
satisfying Ax = b is a convex set. Hints: note that the empty set, ∅, is trivially a convex set
(at least according to me; whether or not our book agrees on this is unclear), and try to use
the previous three problems.
Proof: First, sorry about the misleading hint. You can certainly do it that way, but the
easier way is just to copy the proof from 2:
Choose x1 and x2 that satisfy Ax1 = b and Ax2 = b. Let x lie on the line segment between
x1 and x2 , we want to show that x also satisfies Ax = b. We know that x = λx1 + (1 − λ)x2
for some λ between 0 and 1. Now we have

Ax = A (λx1 + (1 − λ)x2 ) ,
= λAx1 + (1 − λ)Ax2 ,
= λb + (1 − λ)b,
= b,

and this shows that x also lies in this set, proving the claim.

6 Find the extreme points of the set of feasible solutions for the following optimization problem
and then find the optimal solution(s).

Maximize z = 2x + 3y
subject to
3x + y ≤ 6,
x + y ≤ 4,
x + 2y ≤ 6,
x ≥ 0,
y ≥ 0.

Answer: Once again, graphs are a pain, so I’ll skip that part. Here’s a table:

extreme points z = 2x + 3y
(0, 0) 0
(0, 3) 9
(2, 0) 4
( 65 , 12
5
) 48
5
= 9.6

So, the optimal solution is (6/5, 12/5).

You might also like