Emergent Mind

Tight bounds on the expected number of holes in random point sets

(2111.12533)
Published Nov 24, 2021 in math.CO , cs.CG , cs.DM , and math.PR

Abstract

For integers $d \geq 2$ and $k \geq d+1$, a $k$-hole in a set $S$ of points in general position in $\mathbb{R}d$ is a $k$-tuple of points from $S$ in convex position such that the interior of their convex hull does not contain any point from $S$. For a convex body $K \subseteq \mathbb{R}d$ of unit $d$-dimensional volume, we study the expected number $EHK_{d,k}(n)$ of $k$-holes in a set of $n$ points drawn uniformly and independently at random from $K$. We prove an asymptotically tight lower bound on $EHK_{d,k}(n)$ by showing that, for all fixed integers $d \geq 2$ and $k\geq d+1$, the number $EH{d,k}K(n)$ is at least $\Omega(nd)$. For some small holes, we even determine the leading constant $\lim{n \to \infty}n{-d}EHK_{d,k}(n)$ exactly. We improve the currently best known lower bound on $\lim{n \to \infty}n{-d}EHK{d,d+1}(n)$ by Reitzner and Temesvari (2019). In the plane, we show that the constant $\lim{n \to \infty}n{-2}EHK{2,k}(n)$ is independent of $K$ for every fixed $k \geq 3$ and we compute it exactly for $k=4$, improving earlier estimates by Fabila-Monroy, Huemer, and Mitsche (2015) and by the authors (2020).

We're not able to analyze this paper right now due to high demand.

Please check back later (sorry!).

Generate a summary of this paper on our Pro plan:

We ran into a problem analyzing this paper.

Newsletter

Get summaries of trending comp sci papers delivered straight to your inbox:

Unsubscribe anytime.