English

Anticoncentration for subgraph counts in random graphs

Combinatorics 2020-11-19 v2

Abstract

Fix a graph HH and some p(0,1)p\in (0,1), and let XHX_H be the number of copies of HH in a random graph G(n,p)G(n,p). Random variables of this form have been intensively studied since the foundational work of Erd\H{o}s and R\'{e}nyi. There has been a great deal of progress over the years on the large-scale behaviour of XHX_H, but the more challenging problem of understanding the small-ball probabilities has remained poorly understood until now. More precisely, how likely can it be that XHX_H falls in some small interval or is equal to some particular value? In this paper we prove the almost-optimal result that if HH is connected then for any xNx\in \mathbb{N} we have Pr(XH=x)n1v(H)+o(1)\Pr(X_H=x)\le n^{1-v(H)+o(1)}. Our proof proceeds by iteratively breaking XHX_H into different components which fluctuate at "different scales", and relies on a new anticoncentration inequality for random vectors that behave "almost linearly".

Keywords

Cite

@article{arxiv.1905.12749,
  title  = {Anticoncentration for subgraph counts in random graphs},
  author = {Jacob Fox and Matthew Kwan and Lisa Sauermann},
  journal= {arXiv preprint arXiv:1905.12749},
  year   = {2020}
}
R2 v1 2026-06-23T09:32:24.365Z