Scribd
Upload
2 views2 pages

Week 10 Assignment Solutions in Optimization

The document contains solutions to a series of mathematical problems related to dual cones, KKT optimality conditions, tangent cones, linearizing cones, and constraint qualifications. Each problem presents multiple-choice options for the correct answer. The problems cover topics in optimization and convex analysis.

Uploaded by

mdmottahiralam
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
2 views2 pages

Week 10 Assignment Solutions in Optimization

The document contains solutions to a series of mathematical problems related to dual cones, KKT optimality conditions, tangent cones, linearizing cones, and constraint qualifications. Each problem presents multiple-choice options for the correct answer. The problems cover topics in optimization and convex analysis.

Uploaded by

mdmottahiralam
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

Solutions of Assignment of Week 10

Problem 1. Let C = {(x1 , x2 ) ∈ R2 : (x1 + 1)2 + (x2 + 1)2 = 1}. The dual cone of C is

(a) {(x1 , x2 ) ∈ R2 : x1 > 0, x2 > 0}


(b) {(x1 , x2 ) ∈ R2 : x1 > −1, x2 > −1}
(c) {(x1 , x2 ) ∈ R2 : x1 ≤ 0, x2 ≤ 0}
(d) {(x1 , x2 ) ∈ R2 : x1 > 0, x2 < 0}

Problem 2. Karush-Kuhn-Tucker (KKT) optimality conditions for the problem

max f (x)
s.t. g(x) ≤ 0,

where f, g : Rn → R are continuously differentiable functions, are sufficient if

(a) f and g both are concave


(b) f and g both are convex
(c) f is convex and g is concave
(d) f is concave and g is convex

Problem 3. Let Ct (x0 ), Cl (x0 ), and Cf d (x0 ) be tangent cone, linearizing cone, and cone of feasible direc-
tions, respectively, at x0 . Then,

(a) Ct (x0 ) ⊆ Cf d (x0 ) ⊆ Cl (x0 )


(b) Cf d (x0 ) ⊆ Cl (x0 ) ⊆ Ct (x0 )
(c) Cl (x0 ) ⊆ Cf d (x0 ) ⊆ Ct (x0 )
(d) Cf d (x0 ) ⊆ Ct (x0 ) ⊆ Cl (x0 )

Problem 4. The linearizing cone of the set {(x1 , x2 ) ∈ Rn : x2 ≤ x31 , −x2 ≤ 0} at (0, 0) is

(a) {(d1 , d2 ) ∈ R2 : d2 = 0}
(b) {(d1 , d2 ) ∈ R2 : d1 = 0}
(c) {(d1 , d2 ) ∈ R2 : d2 = 0, d1 ≥ 0}
(d) {(d1 , d2 ) ∈ R2 : d1 ≥ 0, d2 ≥ 0}

Problem 5. In constraint qualification, which of the following is true?

(a) ACQ =⇒ MFCQ =⇒ GCQ


(b) MFCQ =⇒ SCQ =⇒ ACQ
(c) ACQ =⇒ MFCQ =⇒ SCQ
(d) SCQ =⇒ MFCQ =⇒ GCQ

Problem 6. Consider the two statements below.

S1: The dual cone of a nonempty set C ⊆ Rn is convex.


S2: The dual cone of a nonempty convex set C ⊆ Rn is convex.

1
Then,
(a) only S1 is true
(b) only S2 is true
(c) both S1 and S2 are true

(d) both S1 and S2 are not true

Problem 7. Which of the following is incorrect?

(a) Origin does not belong to the cone of descent direction.

(b) Origin belongs to the dual cone.


(c) Origin does not belong to the cone of feasible directions.
(d) Origin belongs to the linearizing cone.

Problem 8. A KKT point for the problem


2
min (x1 − 1) + x2
s.t. x1 + x2 ≤ 5
x1 − x2 = −1

is

(a) ( 21 , 23 )
(b) ( 32 , 25 )
(c) (0, 1)

(d) (2, 3)

Problem 9. Let λ be a Lagrange multiplier and (x̄1 , x̄2 , x̄3 , λ) be a stationary point of the Lagrangian
function of the problem
min 2x21 − 24x1 + 2x22 − 8x2 + 2x23 − 12x3 + 200
s.t. x1 + x2 + x3 = 11.
Then, (x̄1 , x̄2 , x̄3 , λ) is

(a) (2, 6, 3, 0)
(b) (3, 6, 2, 0)

(c) (6, 3, 2, 0)
(d) (6, 2, 3, 0)

Problem 10. Let C ∗ denote the dual cone of a convex cone C in Rn . Then, which of the following is true?

(a) C is not necessarily a subset of C ∗∗

(b) C = C ∗∗
(c) C ∗ ⊂ C
(d) C ∗ = C

You might also like