Skip to main content
9 events
when toggle format what by license comment
S Oct 2, 2018 at 18:38 history suggested user127987 CC BY-SA 4.0
corrected notation
Oct 2, 2018 at 14:19 review Suggested edits
S Oct 2, 2018 at 18:38
Jan 23, 2018 at 23:53 comment added Beata Randrianantoanina If you mean how different can the subspaces be that isometric to each other, then I recommend that you look at my paper available on Arxiv math.FA/9709214, where I show an example of two subspaces of L_p, where p\ne 2 is an even integer, so that they are isometric to each other and one subspace is complemented in L_p, while the other one is not complemented. You should check if and how this can be adapted for the finite dimensional case.
Jan 23, 2018 at 20:05 comment added Dave Unfortunately I cannot edit my original question, but I would like to know which subspaces of $\ell_p^n$ are isometric, for $p$ an even integer. My second paragraph describes a special case of the result you just mentioned. To explain: $\ell_p^n$ can be identified as a subspace of $\ell_p^m$, and the subspaces isometric to this subspace are given as the range of $A$, where $A$ is described in my question. What about subspaces that are not spanned by vectors with disjoint support? We can characterise all such isometries for any $p$ that is not an even integer. What is known for even $p$?
Jan 23, 2018 at 18:54 comment added Beata Randrianantoanina I am still not sure what you mean exactly, especially in your second paragraph. Or what do you want your subspaces to be isometric to? A lot is known about isometries of $\ell_p$ and $\ell_p^n$. In particular any subspace spanned by disjointly supported vectors is isometric to $\ell_p$ of appropriate dimension.
Jan 20, 2018 at 14:08 comment added Dave Apologies for my mistake; I mean $\ell_p^n(\mathbb{R})$ instead of $\ell_p^n(\mathbb{R}^n)$. The $n$-dimensional real vector space equipped with the $\ell_p$-norm.
Jan 20, 2018 at 6:20 comment added Beata Randrianantoanina I am somewhat confused by your notation. What do you mean by $\mathbb{R}^n$? Do you mean the $\ell_2$-norm on it? It would be nice if you could fix the various typos in your question. About a reference, I am not sure if this is what you are looking for but in one of my old papers I considered surjective isometries of related spaces, possibly you can deduce something for your situation from it. The paper is available at arxiv.org/pdf/math/9604216.pdf.
Jan 18, 2018 at 23:18 review First posts
Jan 19, 2018 at 1:45
Jan 18, 2018 at 23:17 history asked Dave CC BY-SA 3.0