English

On "stability" in the Erd\H{o}s-Ko-Rado theorem

Combinatorics 2015-02-20 v1

Abstract

Denote by Kp(n,k)K_p(n,k) the random subgraph of the usual Kneser graph K(n,k)K(n,k) in which edges appear independently, each with probability pp. Answering a question of Bollob\'as, Narayanan, and Raigorodskii,we show that there is a fixed p<1p<1 such that a.s. (i.e., with probability tending to 1 as kk \to \infty) the maximum independent sets of Kp(2k+1,k)K_p(2k+1, k) are precisely the sets {AV(K(2k+1,k)):xA}\{A\in V(K(2k+1,k)): x\in A\} (x[2k+1]x\in [2k+1]). We also complete the determination of the order of magnitude of the "threshold" for the above property for general kk and n2k+2n\geq 2k+2. This is new for kn/2k\sim n/2, while for smaller kk it is a recent result of Das and Tran.

Keywords

Cite

@article{arxiv.1502.05692,
  title  = {On "stability" in the Erd\H{o}s-Ko-Rado theorem},
  author = {Pat Devlin and Jeff Kahn},
  journal= {arXiv preprint arXiv:1502.05692},
  year   = {2015}
}

Comments

10 pages