English

The circular law for random regular digraphs with random edge weights

Probability 2017-09-12 v3

Abstract

We consider random n×nn\times n matrices of the form Yn=1dAnXnY_n=\frac1{\sqrt{d}}A_n\circ X_n, where AnA_n is the adjacency matrix of a uniform random dd-regular directed graph on nn vertices, with d=pnd=\lfloor p n\rfloor for some fixed p(0,1)p \in (0,1), and XnX_n is an n×nn\times n matrix of iid centered random variables with unit variance and finite 4+η4+\eta-th moment (here \circ denotes the matrix Hadamard product). We show that as nn\to \infty, the empirical spectral distribution of YnY_n converges weakly in probability to the normalized Lebesgue measure on the unit disk.

Keywords

Cite

@article{arxiv.1508.00208,
  title  = {The circular law for random regular digraphs with random edge weights},
  author = {Nicholas A. Cook},
  journal= {arXiv preprint arXiv:1508.00208},
  year   = {2017}
}
R2 v1 2026-06-22T10:24:22.943Z