Table of Contents
Fetching ...

Ramsey number of a cycle versus a graph of a given size

Stijn Cambie, Andrea Freschi, Patryk Morawski, Kalina Petrova, Alexey Pokrovskiy

TL;DR

The paper addresses the asymmetric Ramsey problem for cycles versus graphs without isolated vertices by proving that for odd $k\ge 7$ and large $e(H)$, $R(C_k,H)\le 2e(H)+\left\lfloor\frac{k-1}{2}\right\rfloor$, settling a long-standing question of Erdős, Faudree, Rousseau and Schelp. The authors develop an induction on $e(H)$, leveraging strong bounds for $R(P_k,H)$ and a neighborhood-decomposition strategy around a vertex with large red degree to partition the host graph into two blue regions connected by blue edges. A key ingredient is a second-neighbourhood analysis: if the second neighbourhood contains a long path, a $C_k$ is forced, enabling the embedding framework. They also provide the exact matching-case bound and extend prior results for even $k$, contributing new techniques that connect path Ramsey bounds with structural neighbourhood arguments. The work advances asymmetric Ramsey theory for cycles and offers tools likely applicable to related cycle-vs-graph Ramsey problems with large edge counts.

Abstract

In this paper, we prove that for every $k$ and every graph $H$ with $m$ edges and no isolated vertices, the Ramsey number $R(C_k,H)$ is at most $2m+\lfloor \frac{k-1}{2} \rfloor$, provided $m$ is sufficiently large with respect to $k$. This settles a problem of Erdős, Faudree, Rousseau and Schelp.

Ramsey number of a cycle versus a graph of a given size

TL;DR

The paper addresses the asymmetric Ramsey problem for cycles versus graphs without isolated vertices by proving that for odd and large , , settling a long-standing question of Erdős, Faudree, Rousseau and Schelp. The authors develop an induction on , leveraging strong bounds for and a neighborhood-decomposition strategy around a vertex with large red degree to partition the host graph into two blue regions connected by blue edges. A key ingredient is a second-neighbourhood analysis: if the second neighbourhood contains a long path, a is forced, enabling the embedding framework. They also provide the exact matching-case bound and extend prior results for even , contributing new techniques that connect path Ramsey bounds with structural neighbourhood arguments. The work advances asymmetric Ramsey theory for cycles and offers tools likely applicable to related cycle-vs-graph Ramsey problems with large edge counts.

Abstract

In this paper, we prove that for every and every graph with edges and no isolated vertices, the Ramsey number is at most , provided is sufficiently large with respect to . This settles a problem of Erdős, Faudree, Rousseau and Schelp.
Paper Structure (3 sections, 9 theorems, 16 equations, 2 figures)

This paper contains 3 sections, 9 theorems, 16 equations, 2 figures.

Key Result

Theorem 1

For any graph $H$ with no isolated vertices, we have $R(K_3, H)\le 2e(H)+1$.

Figures (2)

  • Figure 1: From left to right, the cases $u_j\neq u_{j+k-4}$, $u_1\neq u_2$ and $u_1=u_{k-1}$ in the proof of \ref{['lemma:path_second_neighbourhood']}.
  • Figure 2: In the proof of Theorem \ref{['thm:main+']} we fix a vertex $u$ in our host graph with a large red neighbourhood $U_1$. By Lemma \ref{['lemma:ramsey_of_path_vs_any']} we can embed a large part $H_1$ of $H$ into $U_1$. We can then show that either the second red neighbourhood $\Pi$ of $u$ has size at least $R(P_{2k}, H)$ --- in which case we can find a copy of $H$ in $\Pi$ --- or $U_2 = V(G) \setminus(U_1 \cup \Pi \cup \{u \})$ is large enough for us to find the rest of $H$ in there by induction. Note that by definition all the edges between $U_1$ and $U_2$ are blue.

Theorems & Definitions (19)

  • Theorem 1: Goddard and Kleitman goddard1994upper; Sidorenko Sidorenko93
  • Theorem 3
  • Proposition 4: EFRS78Sudakov02
  • Lemma 5: BEFRS82FL25
  • Lemma 6
  • proof
  • Corollary 7
  • Lemma 8
  • proof
  • Proposition 9: Matching case
  • ...and 9 more