Random Matrices: The circular Law

Abstract

Let $\a$ be a complex random variable with mean zero and bounded variance $σ^{2}$. Let $N_{n}$ be a random matrix of order $n$ with entries being i.i.d. copies of $\a$. Let $λ_{1}, ..., λ_{n}$ be the eigenvalues of $\frac{1}{σ\sqrt n}N_{n}$. Define the empirical spectral distribution $μ_{n}$ of $N_{n}$ by the formula $$ μ_n(s,t) := \frac{1}{n} # \{k \leq n| \Re(λ_k) \leq s; \Im(λ_k) \leq t \}.$$ The Circular law conjecture asserts that $μ_{n}$ converges to the uniform distribution $μ_\infty$ over the unit disk as $n$ tends to infinity. We prove this conjecture under the slightly stronger assumption that the $(2+η)þ$-moment of $\a$ is bounded, for any $η>0$. Our method builds and improves upon earlier work of Girko, Bai, Götze-Tikhomirov, and Pan-Zhou, and also applies for sparse random matrices. The new key ingredient in the paper is a general result about the least singular value of random matrices, which was obtained using tools and ideas from additive combinatorics.