Circular law for random discrete matrices of given row sum

Abstract: Let

M_{n}

be a random matrix of size

n i m e s n

and let

l a m b d a_{1}, ..., l a m b d a_{n}

be the eigenvalues of

M_{n}

. The empirical spectral distribution

m u_{M_{n}}

of

M_{n}

is defined as

mu_{M_n}(s,t)=frac{1}{n}# {kle n, Re(lambda_k)le s; Im(lambda_k)le t}.

The circular law theorem in random matrix theory asserts that if the entries of

M_{n}

are i.i.d. copies of a random variable with mean zero and variance

s i g m a^{2}

, then the empirical spectral distribution of the normalized matrix

f r a c 1 s i g m a s q r t n M_{n}

of

M_{n}

converges almost surely to the uniform distribution

m u_{c} i r

over the unit disk as

n

tends to infinity. In this paper we show that the empirical spectral distribution of the normalized matrix of

M_{n}

, a random matrix whose rows are independent random

(- 1, 1)

vectors of given row-sum

s

with some fixed integer

s

satisfying

| s | l e (1 - o (1)) n

, also obeys the circular law. The key ingredient is a new polynomial estimate on the least singular value of

M_{n}

.

Review text

Consider a random \(n\times n\) matrix whose entries are \(\pm 1\), the rows are independent random vectors and the sum of the elements in each row is \(s_n\), where \(|s_n| < (1-\varepsilon)n\) and \(\varepsilon > 0\) is fixed. The authors prove that the empirical spectral distribution of the matrix \(\frac 1 {\sigma_n \sqrt n}M_n\) with \(\sigma_n^2 = 1- (s_n/n)^2\) converges almost surely to the uniform distribution on the unit disk in the complex plane, as \(n\to\infty\). The key ingredient in the proof is a new polynomial estimate on the least singular value of \(M_n\).

Recommendations

Cited in

(10)

This page was built for publication: Circular law for random discrete matrices of given row sum

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2443433)