On the union of intersecting families

From MaRDI portal
Publication:5222559

DOI10.1017/S096354831900004XzbMATH Open1436.05108arXiv1610.03027MaRDI QIDQ5222559FDOQ5222559


Authors: David Ellis, Noam Lifshitz Edit this on Wikidata


Publication date: 6 April 2020

Published in: Combinatorics, Probability and Computing (Search for Journal in Brave)

Abstract: A family of sets is said to be emph{intersecting} if any two sets in the family have nonempty intersection. In 1973, ErdH{o}s raised the problem of determining the maximum possible size of a union of r different intersecting families of k-element subsets of an n-element set, for each triple of integers (n,k,r). We make progress on this problem, proving that for any fixed integer rgeq2 and for any kleq(frac12o(1))n, if X is an n-element set, and mathcalF=mathcalF1cupmathcalF2cupldotscupmathcalFr, where each mathcalFi is an intersecting family of k-element subsets of X, then |mathcalF|leqnchooseknrchoosek, with equality only if mathcalF=SsubsetX:|S|=k,ScapReqemptyset for some RsubsetX with |R|=r. This is best possible up to the size of the o(1) term, and improves a 1987 result of Frankl and F"uredi, who obtained the same conclusion under the stronger hypothesis k<(3sqrt5)n/2, in the case r=2. Our proof utilises an isoperimetric, influence-based method recently developed by Keller and the authors.


Full work available at URL: https://arxiv.org/abs/1610.03027




Recommendations




Cites Work


Cited In (29)





This page was built for publication: On the union of intersecting families

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5222559)