Alteration method generalization for r̃(m,n)
Informal conjecture on generalization of Theorem 4 to all $\tilde{r}(m,n)$ · arXiv:1806.09726
Status partial medium confidence
The informal conjecture asks whether Krivelevich's generalization of the alteration method to all $r(m,n)$ can be extended to the online setting for all $\tilde{r}(m,n)$. Guo and Warnke (arXiv:1909.02691, published 2023 in Journal of Graph Theory) apply a refined alteration method specifically to online graph Ramsey games, which is directly in the spirit of the conjecture; however, the abstract does not confirm that bounds for all $m$ are established, so a full resolution cannot be verified from available sources.
Cited literature (1)
-
Refines the classical alteration method and applies it to online graph Ramsey games, extending beyond Krivelevich's single-edge-removal approach and likely addressing the conjecture about $\tilde{r}(m,n)$ for general $m$, though the abstract does not specify the exact range of $m$ covered.
Reviewer notes. The conjecture is informal, paralleling Krivelevich's extension of Erdős's alteration method from triangles to general cliques in the classical Ramsey setting. Guo and Warnke (1909.02691) is the most directly relevant follow-up: it refines the alteration method and applies it to online Ramsey games, but the abstract stops short of specifying whether all $m$ are covered. No counterexample or explicit full resolution was found. Xiaoyu He's personal papers page lists no subsequent work on $\tilde{r}(m,n)$.
Context
Theorem 4 establishes $\tilde{r}(3,n) = \Omega(n^3/\log^2 n)$ via a Painter strategy that avoids red triangles, in an argument close in spirit to Erdős's alteration method. The authors note that Krivelevich generalized this alteration method to all $r(m,n)$ and conjecture informally that the same can be done for $\tilde{r}(m,n)$.
Notes. No explicit quantitative bound for general $m$ is stated; only the suspicion of generalizability is expressed.
Source paper
Online Ramsey Numbers and the Subgraph Query Problem
David Conlon, Jacob Fox, Andrey Grinshpun, Xiaoyu He · 2018-11-04
https://arxiv.org/abs/1806.09726
PDF source