2
$\begingroup$

For $n > 1$ an integer, there are well-known formulas for volume of the balls.

What is the analogous statement in a Banach space/Hilbert Space?

$\endgroup$
  • 2
    $\begingroup$ There is no analogue of the volume form for an infinite-dimensional space, since there is no top exterior power. So what does "volume" mean here? $\endgroup$ – Qiaochu Yuan Jul 27 '10 at 21:37
  • $\begingroup$ See the answer of John Cook. $\endgroup$ – user218 Jul 27 '10 at 23:15
4
$\begingroup$

Maybe this is helpful, a formula for the volume of Lp balls in R^n.

$\endgroup$
  • 1
    $\begingroup$ I don't understand how this answers the question. The L^p ball is not the L^p analogue of a simplex. $\endgroup$ – Qiaochu Yuan Jul 28 '10 at 0:00
  • $\begingroup$ @Qiaochu: I was seeding the site with questions and forgot what exact answer I wanted to bait; so I was too vague. Sorry. Now I modified the question. $\endgroup$ – user218 Jul 28 '10 at 13:04
6
$\begingroup$

In an infinite-dimensional normed space with a translation-invariant measure, the measure of a ball must be either $0$ or $\infty$. This fact is sometimes summarized as "there is no infinite-dimensional Lebesgue measure." So unless you have some other notion of "volume" in mind, only the finite-dimensional case (i.e. $\mathbb{R}^n$ with some other norm) has any content. And with regard to Hilbert spaces, the only finite-dimensional Hilbert spaces are $\mathbb{R}^n$ with the Euclidean inner product, so of course we know about this.

$\endgroup$
1
$\begingroup$

A finite measure $\mu$ on a separable complete metric space has the property of tightness; i.e. for any $\epsilon > 0$ there is some compact subset $K$ so that $\mu(K^c) < \epsilon$. A Banach space that is locally compact is necessarily finite dimensional. Hence, a compact subset is a very small subset that must have a void interior.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy