Searcharxiv⌕ Search

arXiv · 2611.16775

Novel upper bounds for Borsuk numbers from contact points and spherical partitions

Abstract

Let $b(n)$ be the least $m$ such that every bounded set in $\mathbb{R}^n$ of positive diameter can be partitioned into $m$ sets of smaller diameter. We prove that \[ b(4)\le6,\qquad b(5)\le12,\qquad b(6)\le31,\qquad b(7)\le45,\qquad b(8)\le89, \] and that \[ b(n)\le1+2^{n-1}\left(\tfrac{39}{64}\right)^{\lfloor(n-1)/8\rfloor} \qquad\text{for every } n\ge9 ; \] for instance, $b(9)\le157$ and $b(17)\le24337$. Further constructions give, for example, $b(10)\le289$ and $b(13)\le2313$. These bounds improve Lassak's bound $b(n)\le2^{n-1}+1$ of 1982 in every dimension $n\ge4$. In dimension four they also improve the recent bounds $b(4)\le8$ (Tolmachev and Voronov, May 2026) and $b(4)\le7$ (Arman, Bondarenko, Prymak and Radchenko, October 2026); the bound $b(4)\le6$ was announced independently by Chalmers in November 2026. The bounds for $n\ge7$ have analytic proofs: a quantitative form of Lassak's reduction to coverings of the sphere is combined with normal partitions of weighted vector configurations, for which the smallest inner product within a region is estimated explicitly. The bound for $n=6$ uses a covering of $S^4$ by thirty sets, checked in rational arithmetic. For $n=4,5$ three contact points of the circumscribed ball reduce the problem to a four-parameter family of bodies, which is split by finitely many power diagrams; that all parts have diameter less than one is verified by a computer program with rigorous error control.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Alexander Tolmachev. 2026-11-12. Novel upper bounds for Borsuk numbers from contact points and spherical partitions. https://searcharxiv.com/abs/2611.16775

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

On the Fourier Mean Bodies of a Convex Body

In 1998, R. Gardner and G. Zhang introduced the radial $p$th mean bodies $R_pK$ of a convex body $K\subset\mathbb R^n$, $p>-1$, which have since become important objects in geometric tomography. In this paper, we study the Fourier transforms of the radial functions of $R_pK$. This leads to a new family of star-shaped sets $F_pK$, which we call the Fourier $p$th mean bodies of $K$. We prove Fourier inversion formulas connecting $R_pK$ and $F_pK$, realizing them as $p$-intersection bodies in the sense of A. Koldobsky. We develop the basic affine geometry of $F_pK$; this includes affine covariance and monotonicity properties. We give sufficient conditions for the compactness of $F_pK$ in terms of the decay of $|\widehat{χ_K}|^2$. We show that $F_pK$ is an origin-symmetric convex body for every $0<p\le1$. This range is sharp in general: already for the cube, $F_p[-1,1]^n$ is not convex for $1<p<2$ and $n\geq 2,$ while $F_p[-1,1]^n$ is not compact for $p\geq 2$. We further investigate the features Fourier mean bodies share with intersection bodies: we prove Hensley-type estimates for $F_pK$ when $K$ is isotropic and investigate an affine isoperimetric inequality.

math.MG↗

Rado's covering problem for cubes and balls: a survey

What is the largest constant $c\in [0,1]$ with the property that every finite collection $\mathcal{C}$ of axis-parallel squares in the plane admits a disjoint sub-collection $\mathcal{S}$ occupying at least a fraction $c$ of the area covered by $\mathcal{C}$? This problem was first raised by T.~Radó in 1928, who was motivated by a classical covering lemma in real analysis due to Vitali. R.~Rado later generalized the problem from axis-parallel squares in the plane to homothetic copies of any given convex body $K$ in $\mathbb{R}^d$, where now we are looking for an optimal constant $F(K)$. In this paper we survey the literature on this problem, with a focus on the case where $K$ is a cube or a ball in the high-dimensional regime $d\rightarrow \infty$. The estimates that we currently have for cubes are much more precise than those for balls: namely if $Q^d$ is a $d$-dimensional cube, then \[ (e^{-1}+o(1))\frac{2^{-d}}{d \log{d}} \leq F(Q^d)\leq 2^{-d}, \] while denoting $B^d$ a $d$-dimensional Euclidean ball, then \[ 2.910^{-d} \leq F(B^d) \leq 2.447^{-d}.\] Except for the upper bound for cubes, which is classical, the other estimates are due to the authors.

math.MG↗

Asymptotic Dimension of Persistence Modules of Bounded Presentation

We compute the asymptotic dimension of spaces of multiparameter persistence modules of bounded presentation under the interleaving distance. The $d$-parameter modules with at most $n$ generators and $m$ relations have asymptotic dimension, and Assouad-Nagata dimension, exactly $d(n+m)$, the number of coordinates of their generator and relation grades, over every field; for two parameters this is $2(n+m)$. The space has continuous moduli, already for two parameters and $n,m\ge2$: over an infinite field it contains, at every scale, infinite uniformly separated families of pairwise non-isomorphic modules with one fibered barcode, so it is not doubling even at a fixed fibered barcode; already for $m\ge1$ it is not coarsely equivalent to a proper metric space. The proof is an induction on the number of distinct grade coordinates: an interleaving small compared with the gaps between these coordinates moves each of them by at most twice its size and preserves the induced grid representation, so the continuous moduli are invisible at large scale. Each class of modules at finite distance therefore embeds coarsely into Hilbert space; for two parameters, $n,m\ge2$ and an uncountable field, a class containing such a family admits no coarse embedding into a separable Hilbert space. Under the two-parameter matching distance, on the space of rank invariants, the asymptotic dimension and the Assouad-Nagata dimension are again $2(n+m)$: slope-one slices pin the grid of grades, and over a grid only finitely many rank invariants occur; on a finite net of $S$ slices the asymptotic dimension is at most $2nS$.

math.MG↗