5
$\begingroup$

Let $\{X_n\}_{n=1}^\infty$ be a sequence of compact Alexandrov spaces (with curvature $\geq k$) converging to (in the sense of Gromov-Hausdroff convergence) an Alexandrov spaces $X$, and $f_n:X_n\mapsto X$ is the corresponding $\nu-$ approximation. The set of regular points in $X_n$ is denoted by $R_n$.

Question: (1) When $n$ sufficiently large, can we prove that $f_n$ is Lipschitz continuous on $R_n$ ? Or a weaker conclusion: $f_n$ is Lipschitz continuous on some $Y_n\subset R_n$ with $\mathrm{vol}(R_n-Y_n)=o(1)$ ,as $n\to\infty$ ?

(2) If question (1) is not true, can we construct a $g_n:X_n\mapsto X$ (maybe related to f_n), such that $g_n$ is Lipschitz continuous on some $Y_n\subset R_n$ with $\mathrm{vol}(R_n-Y_n)=o(1)$ ,as $n\to\infty$ ?

$\endgroup$
4
$\begingroup$

The approximations $f_n$ (by def.) need not to be continuous, but you can approximate Lipschitz ones. I will give a hint how to prove such thing, as far as I know the statement is not written anywhere.

If $X$ is Riemannian the statement follows from Yamaguchi theorem and nearly the same proof works if all the tangent spaces of $X$ are close to Euclidean.

Now assume $X$ has an isolated strongly singular point $p$. Then the above argument gives an approximation which is Lipschitz approximations outside a fixed neighborhood of $p$. Contract all this neighborhood into $p$ and compose the resulting map with the approximation $X_n\to X$. This way you get a Lipschitz approximation.

The general case is none by backward induction on the dimension of strongly singular locus. Section 8 in my paper "Semiconcave functions..." should help. Good luck.

$\endgroup$
  • $\begingroup$ Thank you for your helpful answer ! Can you give some references to the Yamaguchi theorem ? Because I don't know about this. Thank you again. $\endgroup$ – Someone1987 Oct 29 '15 at 9:52
  • $\begingroup$ @Someone1987: Takao Yamaguchi, "A convergence theorem in the geometry of Alexandrov spaces" $\endgroup$ – Anton Petrunin Oct 29 '15 at 13:24

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.