Proving the Containment Property of Polar Cones for Sets in R^n

  • Context: Graduate 
  • Thread starter Thread starter avilaca
  • Start date Start date
  • Tags Tags
    Polar Property
Click For Summary

Discussion Overview

The discussion focuses on the containment property of polar cones for sets in R^n, specifically exploring the relationship between two sets S1 and S2 and their polar cones S1* and S2*. Participants are attempting to prove that if S1 is contained in S2, then S2* is contained in S1*. The conversation also touches on the definition and properties of polar cones, as well as a related challenge involving a specific set defined by a matrix transformation.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant suggests that proving the containment property seems obvious, especially in R^2, but struggles to find a formal proof.
  • Another participant questions whether the proof is evident from the definition based on the inner product.
  • A participant provides a reasoning based on the inner product, stating that if S1 is contained in S2, the region satisfying the condition for S1 is the same or larger than that for S2, implying S2* is contained in S1*.
  • A new challenge is introduced regarding the polar cone of a set defined as S = {x ∈ ℝ^n: x = Ap, p ≥ 0}, prompting further exploration of its polar cone S*.
  • Another participant attempts to define S* using the condition x^T a ≤ 0 for all x in S but expresses difficulty in concluding anything about "a" from this definition.

Areas of Agreement / Disagreement

Participants express differing levels of confidence regarding the proof of the containment property, with some finding it straightforward while others seek further clarification. The discussion regarding the polar cone of the set defined by the matrix transformation remains unresolved, with no consensus on the conclusions about "a".

Contextual Notes

The discussion includes assumptions about the properties of inner products and polar cones, and the reasoning presented relies on these definitions. There are unresolved mathematical steps in deriving the polar cone for the specific set defined by the matrix transformation.

avilaca
Messages
4
Reaction score
0
Let S1*(S2*) be the polar cone of the set S1(S2) (http://en.wikipedia.org/wiki/Dual_cone_and_polar_cone).

How can I show that if S1 is contained in S2 then S2* is contained in S1*.

It looks obvious (especially if we think in R^2), but I do not find a way to prove it.
 
Physics news on Phys.org
welcome to pf!

hi avilaca! welcome to pf! :smile:
avilaca said:
How can I show that if S1 is contained in S2 then S2* is contained in S1*.

It looks obvious (especially if we think in R^2), but I do not find a way to prove it.

isn't the proof obvious from the definition based on inner product? (see http://en.wikipedia.org/wiki/Dual_cone_and_polar_cone) :wink:
 
Ok, it's easy from the inner product <a,s> = ||a||.||s||cos[itex]\theta[/itex].
<a,s> [itex]\leq[/itex] 0 <=> pi/2 [itex]\leq[/itex] [itex]\theta[/itex] [itex]\leq[/itex] 3pi/2.
This means that if S1 [itex]\subset[/itex] S2, by the above result, the region where the condition {<a,s> [itex]\leq[/itex] 0 , s [itex]\in[/itex] S1 or S2, a [itex]\in[/itex] ℝ[itex]^{n}[/itex]} is true for S1 is the same or it's larger than the one for S2, which implies S2* [itex]\subset[/itex] S1*.
 
Now another challenge:

Let S = {x [itex]\in[/itex] ℝ[itex]^{n}[/itex]: x = Ap, p [itex]\geq[/itex] 0}, where A [itex]\in[/itex] M[itex]_{n*m}[/itex], p [itex]\in[/itex] ℝ[itex]^{m}[/itex].
What is its polar cone S*?
 
avilaca said:
Ok, it's easy from the inner product <a,s> = ||a||.||s||cos[itex]\theta[/itex].
<a,s> [itex]\leq[/itex] 0 <=> pi/2 [itex]\leq[/itex] [itex]\theta[/itex] [itex]\leq[/itex] 3pi/2.
This means that if S1 [itex]\subset[/itex] S2, by the above result, the region where the condition {<a,s> [itex]\leq[/itex] 0 , s [itex]\in[/itex] S1 or S2, a [itex]\in[/itex] ℝ[itex]^{n}[/itex]} is true for S1 is the same or it's larger than the one for S2, which implies S2* [itex]\subset[/itex] S1*.

looks good! :smile:
avilaca said:
Now another challenge:

Let S = {x [itex]\in[/itex] ℝ[itex]^{n}[/itex]: x = Ap, p [itex]\geq[/itex] 0}, where A [itex]\in[/itex] M[itex]_{n*m}[/itex], p [itex]\in[/itex] ℝ[itex]^{m}[/itex].
What is its polar cone S*?

show us what you get :wink:
 
I didn't achieve a great conclusion.

S* can be defined by {a [itex]\in[/itex] R[itex]^{n}[/itex]: x[itex]^{T}[/itex]a [itex]\leq[/itex] 0, for all x [itex]\in[/itex] S}.

Now:
x[itex]^{T}[/itex]a [itex]\leq[/itex] 0 <=> (Ap)[itex]^{T}[/itex]a [itex]\leq[/itex] 0 <=> p[itex]^{T}[/itex]A[itex]^{T}[/itex]a [itex]\leq[/itex] 0.

But I can not conclude nothing about "a" from here.
 

Similar threads

Replies
6
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 32 ·
2
Replies
32
Views
4K
  • · Replies 8 ·
Replies
8
Views
4K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 9 ·
Replies
9
Views
2K
Replies
1
Views
5K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K