Abstract
We prove an infinite $(p,q)$-theorem for piercing fat compact convex sets in $\mathbb{R}^d$ with $k$-flats. Additionally, we develop a new framework through which infinite $(p,q)$-theorems concerning compact sets and $k$-flats can be extended to their 'colorful' variants. Further, we show that the existence of an infinite $(p,q)$-theorem does not necessarily imply the existence of the corresponding finite $(p,q)$-theorem.
Introduction
In the field of discrete geometry, Helly's Theorem serves as a foundational result, stating that if every $d+1$ convex sets in $\mathbb{R}^d$ have a non-empty intersection, then the entire collection has a common intersection point. This was significantly generalized by the Alon-Kleitman $(p, q)$-theorem, which establishes that if among every $p$ sets in a family, some $q$ can be pierced by a single point, then the entire family can be pierced by a finite number of points.
However, extending these results to higher-order transversals—where sets are pierced by $k$-dimensional flats (such as lines or planes) rather than points—introduces significant complexity. For general convex sets, $(p, q)$-theorems for $k$-flats often fail to hold. To recover these properties, researchers typically impose "fatness" constraints, ensuring sets are not excessively elongated in any single dimension. This paper addresses the extension of these theorems to infinite families and explores the relationship between monochromatic and "colorful" (heterochromatic) variants.
The Role of Fatness and Convexity
A set is considered $\rho$-fat if it contains a ball of radius $r$ and is contained within a ball of radius $R$, such that the ratio $R/r$ is bounded by $\rho$. The authors demonstrate that fatness is a vital property for higher-order transversals. Without it, even simple configurations in 3D space can fail to satisfy $(p, q)$-style properties.
Furthermore, convexity remains essential. The research highlights that even if sets satisfy fatness requirements, the absence of convexity (for instance, a ball with a thin "tail") can cause the theorems to collapse. By focusing on $\rho$-fat compact convex sets, the authors establish a stable environment for proving higher-order transversal properties.
Infinite $(\aleph_0, q)$ Theorems
The paper introduces an infinite version of the $(p, q)$ property, denoted as $(\aleph_0, q)$. A family of sets satisfies this property if every infinite subset contains at least $q$ sets that can be pierced by a single $k$-flat.
Theorem 2 provides a core result: for a family of $\rho$-fat compact convex sets, if the $(\aleph_0, k+2)$ property holds, then the entire family possesses a finite $k$-transversal. This means the whole collection can be pierced by a finite number of $k$-flats. This result is achieved through induction on both the dimension of the space $d$ and the dimension of the piercing flat $k$.
Heterochromatic Framework
A significant contribution of this work is the development of a "colorful" or heterochromatic framework. In this setting, sets are drawn from different families. The authors provide a logical bridge (Lemma 21) that allows monochromatic theorems (concerning a single family) to be extended to heterochromatic sequences.
Theorem 6 establishes that monochromatic $(\aleph_0, k+2)$-theorems can indeed be extended to these colorful variants. For specific cases like unit balls, Theorem 8 shows that if a sequence of families satisfies a strict heterochromatic property, at least two of those families must have finite point transversals.
The Divergence of Infinite and Finite Theorems
One of the most striking insights of the paper is the demonstration that an infinite $(\aleph_0, q)$-theorem is actually less restrictive than its finite $(p, q)$ counterpart. It was previously an open question whether the existence of an infinite theorem automatically implied the existence of a corresponding finite one.
The authors resolve this with Proposition 9, constructing a collection of $n$ "near-balls" that satisfy the $(q, q)$-property but require $\Theta(n)$ points to pierce globally. This proves that an infinite theorem can exist even in cases where the finite version fails to provide a constant piercing number, showing that the two regimes are fundamentally distinct.
Conclusion and Future Work
This research clarifies the conditions under which higher-order transversals exist for infinite families of convex sets. By establishing the importance of fatness and providing a framework for heterochromatic variants, the authors bridge several gaps in discrete geometry.
However, the relationship between $(\aleph_0, q)$-theorems and finite $(p, q)$-theorems remains a fertile ground for future study. Specifically, determining the exact conditions required to "upgrade" an infinite result to a finite one, and exploring how these properties manifest in other geometric classes—such as axis-parallel boxes—remains an open challenge.
Summary updated