arXiv++ Combinatorics

We consider precolouring extension problems for proper edge-colourings of graphs and multigraphs, in an attempt to prove stronger versions of Vizing's and Shannon's bounds on the chromatic index of (multi)graphs in terms of their maximum degree $Δ$. We are especially interested in the following question: when is it possible to extend a precoloured matching to a colouring of all edges of a (multi)graph? This question turns out to be related to the notorious List Colouring Conjecture and other classic notions of choosability.

A vertex set $D$ in a graph $G$ is $k$-dependent if $G[D]$ has maximum degree at most $k-1$, and $k$-dominating if every vertex outside $D$ has at least $k$ neighbors in $D$. Favaron proved that if $D$ is a $k$-dependent set maximizing the quantity $k|D| - |E(G[D])|$, then $D$ is $k$-dominating. We extend this result, showing that such sets satisfy a stronger structural property, and we find a surprising connection between Favaron's theorem and a conjecture of Tuza regarding packing and covering of triangles.

A graph $G$ is called $(a,b)$-choosable if for any list assignment $L$ which assigns to each vertex $v$ a set $L(v)$ of $a$ permissible colours, there is a $b$-tuple $L$-colouring of $G$. An $(a,1)$-choosable graph is also called $a$-choosable. In the pioneering paper on list colouring of graphs by Erdős, Rubin and Taylor, $2$-choosable graphs are characterized. Confirming a special case of a conjecture of Erdős--Rubin--Taylor, Tuza and Voigt proved that $2$-choosable graphs are $(2m,m)$-choosable for any positive integer $m$. On the other hand, Voigt proved that if $m$ is an odd integer, then these are the only $(2m,m)$-choosable graphs; however, when $m$ is even, there are $(2m,m)$-choosable graphs that are not $2$-choosable. A graph is called $3$-choosable-critical if it is not $2$-choosable, but all its proper subgraphs are $2$-choosable. Voigt conjectured that for every positive integer $m$, all bipartite $3$-choosable-critical graphs are $(4m,2m)$-choosable. In this paper, we determine which $3$-choosable-critical graphs are $(4,2)$-choosable, refuting Voigt's conjecture in the process. Nevertheless, a weaker version of the conjecture is true: we prove that there is an even integer $k$ such that for any positive integer $m$, every bipartite $3$-choosable-critical graph is $(2km,km)$-choosable. Moving beyond $3$-choosable-critical graphs, we present an infinite family of non-$3$-choosable-critical graphs which have been shown by computer analysis to be $(4,2)$-choosable. This shows that the family of all $(4,2)$-choosable graphs has rich structure.

We show that there is a constant $k$ such that when $r \geq 2$ and $m \geq r^k$, the complete $r$-partite graph $K_{m*r}$ has a non-colorable list assignment $L$ such that $|L(v)| \geq \frac{7}{750}r\ln m$ for all $v$ and such that $|L(u) \cap L(v)| \leq \left\lfloor \frac{2r}{r-1} \right\rfloor$ whenever $u \neq v$. This roughly extends a result of Alon to the context of "choosability with separation", introduced by Kratochvíl, Tuza, and Voigt.

Erdős, Gallai, and Tuza posed the following problem: given an $n$-vertex graph $G$, let $τ_1(G)$ denote the smallest size of a set of edges whose deletion makes $G$ triangle-free, and let $α_1(G)$ denote the largest size of a set of edges containing at most one edge from each triangle of $G$. Is it always the case that $α_1(G) + τ_1(G) \leq n^2/4$? We have two main results. We first obtain the upper bound $α_1(G) + τ_1(G) \leq 5n^2/16$, as a partial result towards the Erdős--Gallai--Tuza conjecture. We also show that always $α_1(G) \leq n^2/2 - m$, where $m$ is the number of edges in $G$; this bound is sharp in several notable cases.

Tuza's Conjecture states that if a graph $G$ does not contain more than $k$ edge-disjoint triangles, then some set of at most $2k$ edges meets all triangles of $G$. We prove Tuza's Conjecture for all graphs $G$ having no subgraph with average degree at least $7$. As a key tool in the proof, we introduce a notion of reducible sets for Tuza's Conjecture; these are substructures which cannot occur in a minimal counterexample to Tuza's Conjecture. We also introduce weak König--Egerváry graphs, a generalization of the well-studied König--Egerváry graphs.

We study a game on a graph $G$ played by $r$ {\it revolutionaries} and $s$ {\it spies}. Initially, revolutionaries and then spies occupy vertices. In each subsequent round, each revolutionary may move to a neighboring vertex or not move, and then each spy has the same option. The revolutionaries win if $m$ of them meet at some vertex having no spy (at the end of a round); the spies win if they can avoid this forever. Let $σ(G,m,r)$ denote the minimum number of spies needed to win. To avoid degenerate cases, assume $|V(G)|\ge r-m+1\ge\floor{r/m}\ge 1$. The easy bounds are then $\floor{r/m}\le σ(G,m,r)\le r-m+1$. We prove that the lower bound is sharp when $G$ has a rooted spanning tree $T$ such that every edge of $G$ not in $T$ joins two vertices having the same parent in $T$. As a consequence, $σ(G,m,r)\leγ(G)\floor{r/m}$, where $γ(G)$ is the domination number; this bound is nearly sharp when $γ(G)\le m$. For the random graph with constant edge-probability $p$, we obtain constants $c$ and $c'$ (depending on $m$ and $p$) such that $σ(G,m,r)$ is near the trivial upper bound when $r<c\ln n$ and at most $c'$ times the trivial lower bound when $r>c'\ln n$. For the hypercube $Q_d$ with $d\ge r$, we have $σ(G,m,r)=r-m+1$ when $m=2$, and for $m\ge 3$ at least $r-39m$ spies are needed. For complete $k$-partite graphs with partite sets of size at least $2r$, the leading term in $σ(G,m,r)$ is approximately $\frac{k}{k-1}\frac{r}{m}$ when $k\ge m$. For $k=2$, we have $σ(G,2,r)=\bigl\lceil{\frac{\floor{7r/2}-3}5}\bigr\rceil$ and $σ(G,3,r)=\floor{r/2}$, and in general $\frac{3r}{2m}-3\le σ(G,m,r)\le\frac{(1+1/\sqrt3)r}{m}$.

We study the Maker-Breaker game on the hypergraph of chains of fixed size in a poset. In a product of chains, the maximum size of a chain that Maker can guarantee building is $k-\lfloor r/2\rfloor$, where $k$ is the maximum size of a chain in the product, and $r$ is the maximum size of a factor chain. We also study a variant in which Maker must follow the chain in order, called the {\it Walker-Blocker game}. In the poset consisting of the bottom $k$ levels of the product of $d$ arbitrarily long chains, Walker can guarantee a chain that hits all levels if $d\ge14$; this result uses a solution to Conway's Angel-Devil game. When d=2, the maximum that Walker can guarantee is only 2/3 of the levels, and 2/3 is asymptotically achievable in the product of two equal chains.