Does Tverberg's theorem hold for CAT(0) spaces of covering dimension $d<\infty$:

Is it true that for any $d$-dimensional $CAT(0)$-space $X$ and a subset $E\subset X$ of cardinality $(d + 1)(r - 1) + 1$, there exists a point $x\in X$ and a partition of $E$ into $r$ subsets $E_1,...,E_r$, such that $x$ belongs to the intersection of closed convex hulls of the subsets $E_i$?


1 Answer 1


No. Let $X$ be a tripod (three segments with one common endpoint), $d=1$, $r=2$ and $E$ the set of the 3 leaf points.

  • $\begingroup$ But it is okey for 4 points... $\endgroup$ Nov 29, 2013 at 23:57
  • $\begingroup$ It seems that $2^{n+1}$ points in $n$-dimensional CAT(0) space can be always divided in two groups. $\endgroup$ Nov 30, 2013 at 0:08
  • $\begingroup$ So, this might parallel the situation with "topological Tverberg's theorem". $\endgroup$
    – Misha
    Nov 30, 2013 at 7:15
  • 2
    $\begingroup$ In a Hadamard $d$-manifold, $d+2$ points seems to be enough for $r=2$. This follows by applying Borsuk-Ulam theorem to a suitable barycentric map. $\endgroup$ Dec 1, 2013 at 9:53
  • 1
    $\begingroup$ non-manifold case is different. However, it would be interesting to figure out what the right bound is in the pure CAT(0) case. At the moment, I do not have a conjecture. Here is the reference to topological Tverberg: renyi.hu/~barany/cikkek/5.pdf. $\endgroup$
    – Misha
    Dec 1, 2013 at 21:32

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.