Is Every Norm on a Finite-Dimensional Vector Space Induced by an Inner Product?

Click For Summary

Discussion Overview

The discussion centers around whether every norm on a finite-dimensional vector space over the real or complex numbers is induced by an inner product. Participants explore implications of this question, particularly in relation to the weak closure of the unit sphere in finite-dimensional normed vector spaces.

Discussion Character

  • Debate/contested
  • Technical explanation
  • Conceptual clarification

Main Points Raised

  • Some participants assert that not every norm is induced by an inner product, citing the failure of the parallelogram law for various norms, such as all p-norms except when p=2.
  • Others argue that in finite-dimensional vector spaces, all norm topologies are equivalent, suggesting that one can assume a norm topology arises from a norm induced by an inner product.
  • It is noted that if a norm satisfies the parallelogram law, an inner product can be derived from the norm using the polarization identity.
  • A clarification is provided regarding the polarization identity, indicating that its form differs when working over the real numbers versus the complex numbers.

Areas of Agreement / Disagreement

Participants express differing views on whether every norm is induced by an inner product, with some asserting it is not possible while others suggest that in finite dimensions, norms can be equivalent to those induced by inner products. The discussion remains unresolved regarding the generality of the claim.

Contextual Notes

The discussion highlights the dependence on specific properties of norms, such as the parallelogram law, and the implications of dimensionality on the relationships between norms and inner products. There are also distinctions made between real and complex vector spaces that may affect the application of the polarization identity.

owlpride
Messages
210
Reaction score
0
On a finite-dimensional vector space over R or C, is every norm induced by an inner product?

I know that this can fail for infinite-dimensional vector spaces. It just struck me that we never made a distinction between normed vector spaces and inner product spaces in my linear algebra course on finite-dimensional vector spaces.

Why I actually care about it: I wonder why the unit sphere in a finite-dimensional normed vector space is weakly closed. Obviously the statement should be that a sequence in a finite-dimensional space converges weakly if and only if it converges strongly, but I'm not sure how to go about this without using an inner product.
 
Physics news on Phys.org
owlpride said:
On a finite-dimensional vector space over R or C, is every norm induced by an inner product?
No. The parallelogram law can and does fail for a variety of norms (e.g. all the p-norms except when p=2).

Why I actually care about it: I wonder why the unit sphere in a finite-dimensional normed vector space is weakly closed. Obviously the statement should be that a sequence in a finite-dimensional space converges weakly if and only if it converges strongly, but I'm not sure how to go about this without using an inner product.
On a finite-dimensional vector space, all norm topologies are equivalent. So you can always assume your norm topology comes from a norm that is induced by an inner product. (In fact it's also true that in this case the weak topology and any norm topology are the same.)
 
It should be noted that if a norm satisfies the parallelogram law, then you can always uncover an inner product from the norm via the "polarization identity"

1/4 ( | v + w |^2 - | v - w |^2 ) = < v , w >
 
wisvuze said:
It should be noted that if a norm satisfies the parallelogram law, then you can always uncover an inner product from the norm via the "polarization identity"

1/4 ( | v + w |^2 - | v - w |^2 ) = < v , w >
Just for the sake of completeness: that form of the polarization identity is only valid if we're working over R; over C, the identity becomes $$\frac14 (\|v+w\|^2 -\|v-w\|^2 + i\|v+iw\|^2 - i\|v-iw\|^2) = \langle v, w\rangle.$$
 

Similar threads

  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 13 ·
Replies
13
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 15 ·
Replies
15
Views
11K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K