Papers
Topics
Authors
Recent
Search
2000 character limit reached

A generalization of the Kővári-Sós-Turán theorem

Published 13 Feb 2020 in math.CO and cs.DM | (2002.05336v2)

Abstract: We present a new proof of the K\H{o}v\'{a}ri-S\'{o}s-Tur\'{a}n theorem that ex(n,Ks,t)=O(n<sup>21/t)ex(n, K_{s,t}) = O(n<sup>{2-1/t}) for s,t2s, t \geq 2. The new proof is elementary, avoiding the use of convexity. For any dd-uniform hypergraph HH, let exd(n,H)ex_d(n,H) be the maximum possible number of edges in an HH-free dd-uniform hypergraph on nn vertices. Let KH,tK_{H, t} be the (d+1)(d+1)-uniform hypergraph obtained from HH by adding tt new vertices v1,,vtv_1, \dots, v_t and replacing every edge ee in E(H)E(H) with tt edges $e \cup \left{v_1\right},\dots, e \cup \left{v_t\right}$ in E(KH,t)E(K_{H, t}). If HH is the $1$-uniform hypergraph on ss vertices with ss edges, then KH,t=Ks,tK_{H, t} = K_{s, t}. We prove that exd+1(n,KH,t)=O(exd(n,H)<sup>1/t</sup>n<sup>d+1d/t</sup>+tn<sup>d)ex_{d+1}(n,K_{H,t}) = O(ex_d(n, H)<sup>{1/t}</sup> n<sup>{d+1-d/t}</sup> + t n<sup>d) for any dd-uniform hypergraph HH with at least two edges such that exd(n,H)=o(n<sup>d)ex_d(n, H) = o(n<sup>d). Thus exd+1(n,KH,t)=O(n<sup>d+11/t)ex_{d+1}(n,K_{H,t}) = O(n<sup>{d+1-1/t}) for any dd-uniform hypergraph HH with at least two edges such that exd(n,H)=O(n<sup>d1)ex_d(n, H) = O(n<sup>{d-1}), which implies the K\H{o}v\'{a}ri-S\'{o}s-Tur\'{a}n theorem in the d=1d = 1 case. This also implies that exd+1(n,KH,t)=O(n<sup>d+11/t)ex_{d+1}(n, K_{H,t}) = O(n<sup>{d+1-1/t}) when HH is a dd-uniform hypergraph with at least two edges in which all edges are pairwise disjoint, which generalizes an upper bound proved by Mubayi and Verstra\"{e}te (JCTA, 2004). We also obtain analogous bounds for 0-1 matrix Tur\'{a}n problems.

Authors (1)

Summary

No one has generated a summary of this paper yet.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.