English

Rank of Sparse Bernoulli Matrices

Probability 2025-05-20 v3

Abstract

Let An A_n be an n×nn \times n random matrix with i.i.d Bernoulli(pp) entries. For a fixed positive integer β\beta, suppose pp satisfies log(n)npcβ \frac{ \log(n) }{ n } \le p \le c_\beta where cβ(0,1/2)c_\beta \in ( 0, 1/2 ) is a β\beta-dependentvalue. For t0t \ge 0, \mathbb{P} \left\{ s_{ n - \beta + 1}(A) \le t n^{-2\beta + \mathfrak{n}(1) }(pn)^{-7} \right\} = t + ( 1 + o_\mathfrak{n}(1) ) \mathbb{P} \bigg\{ \mbox{either $\beta$ rows or $\beta$ columns of $A_n$ equal $\vec{0}$} \bigg\}.

Keywords

Cite

@article{arxiv.2009.13726,
  title  = {Rank of Sparse Bernoulli Matrices},
  author = {Han Huang},
  journal= {arXiv preprint arXiv:2009.13726},
  year   = {2025}
}

Comments

revision

R2 v1 2026-06-23T18:51:56.634Z