Christophe Paul,Evangelos Protopapas,Dimitrios M. Thilikos,Sebastian Wiederrecht
Abstract:Let ${\cal G}$ and ${\cal H}$ be minor-closed graph classes. The pair $({\cal H},{\cal G})$ is an Erdős-Pósa pair (EP-pair) if there is a function $f$ where, for every $k$ and every $G\in{\cal G},$ either $G$ has $k$ pairwise vertex-disjoint subgraphs not belonging to ${\cal H},$ or there is a set $S\subseteq V(G)$ where $|S|\leq f(k)$ and $G-S\in{\cal H}.$ The classic result of Erdős and Pósa says that if $\mathcal{F}$ is the class of forests, then $({\cal F},{\cal G})$ is an EP-pair for every ${\cal G}$. The class ${\cal G}$ is an EP-counterexample for ${\cal H}$ if ${\cal G}$ is minimal with the property that $({\cal H},{\cal G})$ is not an EP-pair. We prove that for every ${\cal H}$ the set $\mathfrak{C}_{\cal H}$ of all EP-counterexamples for ${\cal H}$ is finite. In particular, we provide a complete characterization of $\mathfrak{C}_{\cal H}$ for every ${\cal H}$ and give a constructive upper bound on its size. Each class ${\cal G}\in \mathfrak{C}_{\cal H}$ can be described as all minors of a sequence of grid-like graphs $\langle \mathscr{W}_{k} \rangle_{k\in \mathbb{N}}.$ Moreover, each $\mathscr{W}_{k}$ admits a half-integral packing: $k$ copies of some $H\not\in{\cal H}$ where no vertex is used more than twice. This gives a complete delineation of the half-integrality threshold of the Erdős-Pósa property for minors and yields a constructive proof of Thomas' conjecture on the half-integral Erdős-Pósa property for minors (recently confirmed, non-constructively, by Liu). Let $h$ be the maximum size of a graph in ${\cal H}.$ For every class ${\cal H},$ we construct an algorithm that, given a graph $G$ and a $k,$ either outputs a half-integral packing of $k$ copies of some $H \not\in {\cal H}$ or outputs a set of at most ${2^{k^{\cal O}_h(1)}}$ vertices whose deletion creates a graph in ${\cal H}$ in time $2^{2^{k^{{\cal O}_h(1)}}}\cdot |G|^4\log |G|.$
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the obstruction problem regarding the Erdős - Pósa property in graph theory. Specifically, the authors are concerned with determining which graph classes $\mathcal{H}$ and ambient graph classes $\mathcal{G}$ form an Erdős - Pósa pair (EP - pair), and when these graph classes do not form an EP - pair, how to describe these counterexamples.
### Background and Motivation
The Erdős - Pósa property is an important concept in graph theory, which describes the relationship between finding several pairwise vertex - disjoint subgraphs (called "packing") in a graph and finding a smaller set of vertices such that after deleting these vertices, the graph no longer contains these subgraphs (called "hitting set"). The classical Erdős - Pósa theorem states that for the forest class $\mathcal{F}$, any graph class $\mathcal{G}$ forms an EP - pair with $\mathcal{F}$.
However, for more general graph classes $\mathcal{H}$, this property does not always hold. Therefore, studying which graph classes $\mathcal{H}$ and ambient graph classes $\mathcal{G}$ form an EP - pair, and when they do not form an EP - pair, how to describe these counterexamples, is an important problem.
### Main Contributions
1. **Finiteness**: The authors prove that for every minor - closed graph class $\mathcal{H}$, the set $\mathcal{C}_\mathcal{H}$ of all EP - counterexamples is finite. This means that there exists a finite set of graph classes, which are the minimal counterexamples, that is, they do not form an EP - pair with $\mathcal{H}$, but any graph class properly contained in them forms an EP - pair with $\mathcal{H}$.
2. **Constructive Description**: The authors provide a complete constructive description of $\mathcal{C}_\mathcal{H}$. Specifically, each EP - counterexample can be described as the set of all subgraphs of certain grid - like graphs (called "walloids"). These grid - like graphs have semi - integral $\mathcal{H}$ - obstructions, that is, each vertex is used at most twice.
3. **Algorithmic Results**: The authors also provide an algorithm. Given a graph $G$ and an integer $k$, the algorithm either outputs a set of semi - integral $k$ $\mathcal{H}$ - obstructions or outputs a set of vertices with a size not exceeding $2k\cdot O(h^1)$. After deleting these vertices, the graph $G$ belongs to $\mathcal{H}$. Here, $h = h(\mathcal{H})$ represents the maximum obstruction size of $\mathcal{H}$.
4. **Parameterized Graphs**: The authors introduce the concept of parameterized graphs and construct a finite set $\mathcal{F}_{1/2}^\mathcal{H}$ of parameterized graphs, which can be used to describe all EP - counterexamples. Each parameterized graph $\langle W_t \rangle_{t\in\mathbb{N}}$ is a "grid - like" sequence of graphs. As $t$ increases, these graphs require larger $\mathcal{H}$ - obstructions.
### Conclusion
Through these results, the authors not only solve the obstruction problem of the Erdős - Pósa property but also provide a constructive method to describe these obstructions. This provides new tools and theoretical foundations for many algorithmic applications in graph theory. Moreover, these results generalize Robertson and Seymour's Grid Theorem, further enriching the structural theory in graph theory.