A Proletarian Approach to Generating Eigenvalues of GUE Matrices

Luc Devroye,Jad Hamdan

2023-09-14

Abstract:We propose a simple algorithm to generate random variables described by densities equaling squared Hermite functions. Using results from random matrix theory, we utilize this to generate a randomly chosen eigenvalue of a matrix from the Gaussian Unitary Ensemble (GUE) in sublinear expected time in the RAM model.

Probability

What problem does this paper attempt to address?

Distribution of the number of pivots needed using Gaussian elimination with partial pivoting on random matrices

John Peca-Medlin

DOI: https://doi.org/10.1214/23-AAP2023

2023-09-14

Abstract:Gaussian elimination with partial pivoting (GEPP) is a widely used method to solve dense linear systems. Each GEPP step uses a row transposition pivot movement if needed to ensure the leading pivot entry is maximal in magnitude for the leading column of the remaining untriangularized subsystem. We will use theoretical and numerical approaches to study how often this pivot movement is needed. We provide full distributional descriptions for the number of pivot movements needed using GEPP using particular Haar random ensembles, as well as compare these models to other common transformations from randomized numerical linear algebra. Additionally, we introduce new random ensembles with fixed pivot movement counts and fixed sparsity, $\alpha$. Experiments estimating the empirical spectral density (ESD) of these random ensembles leads to a new conjecture on a universality class of random matrices with fixed sparsity whose scaled ESD converges to a measure on the complex unit disk that depends on $\alpha$ and is an interpolation of the uniform measure on the unit disk and the Dirac measure at the origin.

Numerical Analysis,Probability
Extreme eigenvalues of Laplacian random matrices with Gaussian entries

Andrew Campbell,Kyle Luh,Sean O'Rourke,Santiago Arenas-Velilla,Victor Pérez-Abreu

2024-08-20

Abstract:A Laplacian matrix is a real symmetric matrix whose row and column sums are zero. We investigate the limiting distribution of the largest eigenvalues of a Laplacian random matrix with Gaussian entries. Unlike many classical matrix ensembles, this random matrix model contains dependent entries. Our main results show that the extreme eigenvalues of this model exhibit Poisson statistics. In particular, after properly shifting and scaling, we show that the largest eigenvalue converges to the Gumbel distribution as the dimension of the matrix tends to infinity. While the largest diagonal entry is also shown to have Gumbel fluctuations, there is a rather surprising difference between its deterministic centering term and the centering term required for the largest eigenvalues.

Probability
PDEs satisfied by extreme eigenvalues distributions of GUE and LUE

Estelle Basor,Yang Chen,Lun Zhang

DOI: https://doi.org/10.48550/arXiv.1102.0402

2011-02-02

Abstract:In this paper we study, $\textsf{Prob}(n,a,b),$ the probability that all the eigenvalues of finite $n$ unitary ensembles lie in the interval $(a,b)$. This is identical to the probability that the largest eigenvalue is less than $b$ and the smallest eigenvalue is greater than $a$. It is shown that a quantity allied to $\textsf{Prob}(n,a,b)$, namely, $$ H_n(a,b):=\left[\frac{\partial}{\partial a}+\frac{\partial}{\partial b}\right]\ln\textsf{Prob}(n,a,b),$$ in the Gaussian Unitary Ensemble (GUE) and $$ H_n(a,b):=\left[a\frac{\partial}{\partial a}+b\frac{\partial}{\partial b}\right]\ln \textsf{Prob}(n,a,b),$$ in the Laguerre Unitary Ensemble (LUE) satisfy certain nonlinear partial differential equations for fixed $n$, interpreting $H_n(a,b)$ as a function of $a$ and $b$. These partial differential equations maybe considered as two variable generalizations of a Painlevé IV and a Painlevé V system, respectively. As an application of our result, we give an analytic proof that the extreme eigenvalues of the GUE and the LUE, when suitably centered and scaled, are asymptotically independent.

Mathematical Physics,Probability
Asymptotic Gap Probability Distributions of the Gaussian Unitary Ensembles and Jacobi Unitary Ensembles

Shulin Lyu,Yang Chen,Engui Fan

DOI: https://doi.org/10.1016/j.nuclphysb.2017.11.018

IF: 3.045

2017-01-01

Nuclear Physics B

Abstract:In this paper, we address a class of problems in unitary ensembles. Specifically, we study the probability that a gap symmetric about 0, i.e. (−a,a) is found in the Gaussian unitary ensembles (GUE) and the Jacobi unitary ensembles (JUE) (where in the JUE, we take the parameters α=β). By exploiting the even parity of the weight, a doubling of the interval to (a2,∞) for the GUE, and (a2,1), for the (symmetric) JUE, shows that the gap probabilities maybe determined as the product of the smallest eigenvalue distributions of the LUE with parameter α=−1/2, and α=1/2 and the (shifted) JUE with weights x1/2(1−x)β and x−1/2(1−x)β. The σ function, namely, the derivative of the log of the smallest eigenvalue distributions of the finite-n LUE or the JUE, satisfies the Jimbo–Miwa–Okamoto σ form of PV and PVI, although in the shift Jacobi case, with the weight xα(1−x)β, the β parameter does not show up in the equation. We also obtain the asymptotic expansions for the smallest eigenvalue distributions of the Laguerre unitary and Jacobi unitary ensembles after appropriate double scalings, and obtained the constants in the asymptotic expansion of the gap probabilities, expressed in term of the Barnes G-function valuated at special point.
Strong convergence for tensor GUE random matrices

Benoît Collins,Wangjun Yuan

2024-07-12

Abstract:Haagerup and Thorbjørnsen proved that iid GUEs converge strongly to free semicircular elements as the dimension grows to infinity. Motivated by considerations from quantum physics -- in particular, understanding nearest neighbor interactions in quantum spin systems -- we consider iid GUE acting on multipartite state spaces, with a mixing component on some sites and identity on the remaining sites. We show that under proper assumptions on the dimension of the sites, strong asymptotic freeness still holds. Our proof relies on an interpolation technology recently introduced by Bandeira, Boedihardjo and van Handel.

Probability,Mathematical Physics,Operator Algebras
The Exact Density of the Eigenvalues of the Wishart and Matrix-Variate Gamma and Beta Random Variables

A. M. Mathai,Serge B. Provost

DOI: https://doi.org/10.3390/math12152427

IF: 2.4

2024-08-06

Mathematics

Abstract:The determination of the distributions of the eigenvalues associated with matrix-variate gamma and beta random variables of either type proves to be a challenging problem. Several of the approaches utilized so far yield unwieldy representations that, for instance, are expressed in terms of multiple integrals, functions of skew symmetric matrices, ratios of determinants, solutions of differential equations, zonal polynomials, and products of incomplete gamma or beta functions. In the present paper, representations of the density functions of the smallest, largest and jth largest eigenvalues of matrix-variate gamma and each type of beta random variables are explicitly provided as finite sums when certain parameters are integers and, as explicit series, in the general situations. In each instance, both the real and complex cases are considered. The derivations initially involve an orthonormal or unitary transformation whereby the wedge products of the differential elements of the eigenvalues can be worked out from those of the original matrix-variate random variables. Some of these results also address the distribution of the eigenvalues of a central Wishart matrix as well as eigenvalue problems arising in connection with the analysis of variance procedure and certain tests of hypotheses in multivariate analysis. Additionally, three numerical examples are provided for illustration purposes.

mathematics
Computing marginal eigenvalue distributions for the Gaussian and Laguerre orthogonal ensembles

Peter J. Forrester,Santosh Kumar,Bo-Jian Shen

2024-11-24

Abstract:The Gaussian and Laguerre orthogonal ensembles are fundamental to random matrix theory, and the marginal eigenvalue distributions are basic observable quantities. Notwithstanding a long history, a formulation providing high precision numerical evaluations for $N$ large enough to probe asymptotic regimes, has not been provided. An exception is for the largest eigenvalue, where there is a formalism due to Chiani which uses a combination of the Pfaffian structure underlying the ensembles, and a recursive computation of the matrix elements. We augment this strategy by introducing a generating function for the conditioned gap probabilities. A finite Fourier series approach is then used to extract the sequence of marginal eigenvalue distributions as a linear combination of Pfaffians, with the latter then evaluated using an efficient numerical procedure available in the literature. Applications are given to illustrating various asymptotic formulas, local central limit theorems, and central limit theorems, as well as to probing finite size corrections. Further, our data indicates that the mean values of the marginal distributions interlace with the zeros of the Hermite polynomial (Gaussian ensemble) and a Laguerre polynomial (Laguerre ensemble).

Mathematical Physics,Statistics Theory
Universal cubic eigenvalue repulsion for random normal matrices

Gary Oas

DOI: https://doi.org/10.1103/physreve.55.205

IF: 2.707

1997-01-01

Physical Review E

Abstract:Random matrix models consisting of normal matrices, defined by the sole constraint [N†,N]=0, will be explored. It is shown that cubic eigenvalue repulsion in the complex plane is universal with respect to the probability distribution of matrices. The density of eigenvalues, all correlation functions, and level spacing statistics are calculated. Normal matrix models offer more probability distributions amenable to analytical analysis than complex matrix models where only a model with a Gaussian distribution is solvable. The statistics of numerically generated eigenvalues from Gaussian distributed normal matrices are compared to the analytic results obtained and agreement is seen.

physics, fluids & plasmas, mathematical
A new approach to the characteristic polynomial of a random unitary matrix

Yacine Barhoumi-Andréani

DOI: https://doi.org/10.48550/arXiv.2011.02465

2020-11-05

Abstract:Since the seminal work of Keating and Snaith, the characteristic polynomial of a random Haar-distributed unitary matrix has seen several of its functional studied or turned into a conjecture; for instance: $ \bullet $ its value in $1$ (Keating-Snaith theorem), $ \bullet $ the truncation of its Fourier series up to any fraction of its degree, $ \bullet $ the computation of the relative volume of the Birkhoff polytope, $ \bullet $ its products and ratios taken in different points, $ \bullet $ the product of its iterated derivatives in different points, $ \bullet $ functionals in relation with sums of divisor functions in $ \mathbb{F}_q[X] $. $ \bullet $ its mid-secular coefficients, $ \bullet $ the "moments of moments", etc. We revisit or compute for the first time the asymptotics of the integer moments of these last functionals and several others. The method we use is a very general one based on reproducing kernels, a symmetric function generalisation of some classical orthogonal polynomials interpreted as the Fourier transform of particular random variables and a local Central Limit Theorem for these random variables. We moreover provide an equivalent paradigm based on a randomisation of the mid-secular coefficients to rederive them all. These methodologies give a new and unified framework for all the considered limits and explain the apparition of Hankel determinants or Wronskians in the limiting functional.

Probability
Eigenvalues of random matrices from compact classical groups in Wasserstein metric

Bence Borda

2023-11-15

Abstract:The circular unitary ensemble and its generalizations concern a random matrix from a compact classical group $\mathrm{U}(N)$, $\mathrm{SU}(N)$, $\mathrm{O}(N)$, $\mathrm{SO}(N)$ or $\mathrm{USp}(N)$ distributed according to the Haar measure. The eigenvalues are known to be very evenly distributed on the unit circle. In this paper, we study the distance from the empirical measure of the eigenvalues to uniformity in the quadratic Wasserstein metric $W_2$. After finding the exact value of the expected value and the variance, we deduce a limit law for the square of the Wasserstein distance. We reformulate our results in terms of the $L^2$ average of the number of eigenvalues in circular arcs, and also in terms of the characteristic polynomial of the matrix on the unit circle.

Probability
On Random Matrices Arising in Deep Neural Networks. Gaussian Case

Leonid Pastur

DOI: https://doi.org/10.48550/arXiv.2001.06188

2020-04-07

Abstract:The paper deals with distribution of singular values of product of random matrices arising in the analysis of deep neural networks. The matrices resemble the product analogs of the sample covariance matrices, however, an important difference is that the population covariance matrices, which are assumed to be non-random in the standard setting of statistics and random matrix theory, are now random, moreover, are certain functions of random data matrices. The problem has been considered in recent work [21] by using the techniques of free probability theory. Since, however, free probability theory deals with population matrices which are independent of the data matrices, its applicability in this case requires an additional justification. We present this justification by using a version of the standard techniques of random matrix theory under the assumption that the entries of data matrices are independent Gaussian random variables. In the subsequent paper [18] we extend our results to the case where the entries of data matrices are just independent identically distributed random variables with several finite moments. This, in particular, extends the property of the so-called macroscopic universality on the considered random matrices.

Mathematical Physics,Machine Learning
Asymptotic free independence and entry permutations for Gaussian random matrices. Part II: Infinitesimal freeness

M. Popa,K. Szpojankowski,P.-L. Tseng

2024-09-27

Abstract:We study asymptotic infinitesimal distributions of Gaussian Unitary Ensembles with permuted entries. We show that for random uniform permutations, the asymptotically permuted GUE matrix has a null infinitesimal distribution. Moreover, we show that asymptotically different permutations of the same GUE matrix are infinitesimally free. Besides this we study particular example of entry permutation - the transpose, and we show that while a GUE matrix is asymptotically free from its transpose it is not infinitesimally free from it.

Probability,Combinatorics,Operator Algebras
Combinatorial aspects of Parraud's asymptotic expansion for GUE matrices

David Jekel

2024-02-13

Abstract:We give a new combinatorial proof of Parraud's formula for the asymptotic expansion in powers of $1/N^2$ for the expected trace of polynomials of several independent $N \times N$ GUE matrices, which expresses the result using a mixture of free difference quotients, introducing new freely independent semicircular variables, and integration with respect to parameters. Our approach streamlines the statement of the formula while clarifying its relationship to the combinatorial genus expansion.

Combinatorics,Operator Algebras,Probability
Edge universality of sparse Erdős-Rényi digraphs

Yukun He

2024-09-21

Abstract:Let $\mathcal A$ be the adjacency matrix of the Erdős-Rényi directed graph $\mathscr G(N,p)$. We denote the eigenvalues of $\mathcal A$ by $\lambda_1^{\cal A},...,\lambda^{\cal A}_N$, and $|\lambda_1^{\cal A}|=\max_i|\lambda_i^{\cal A}|$. For $N^{-1+o(1)}\leq p\leq 1/2$, we show that \[ \max_{i=2,3,...,N} \bigg|\frac{\lambda_i^{\mathcal A}}{\sqrt{Np(1-p)}}\bigg| =1+O(N^{-1/2+o(1)}) \] with very high probability. In addition, we prove that near the unit circle, the local eigenvalue statistics of ${\mathcal A}/\sqrt{Np(1-p)}$ coincide with those of the real Ginibre ensemble. As a by-product, we also show that all non-trivial eigenvectors of $\mathcal A$ are completely delocalized. For Hermitian random matrices, it is known that the edge statistics are sensitive to the sparsity: in the very sparse regime, one needs to remove many noise random variables (which affect both the mean and the fluctuation) to recover the Tracy-Widom distribution. Our results imply that, compared to their analogues in the Hermitian case, the edge statistics of non-Hermitian sparse random matrices are more robust. The edge of non-Hermitian matrices possesses the cusp singularity, which was believed to be a technical difficulty of the model. Our first novelty is the observation that the cusp singularity is in fact an advantage instead of an obstacle, and when used properly, it can make the computation easier for non-Hermitian matrices. The second novelty is an use of integration by parts formula inside the Girko's Hermitization, which completely avoids the study of the Green function at larger spectral scales. The third novelty is the self-similarity of the self-consistent equations of certain Green functions, which eliminates the effect of large expectations of the matrix entries.

Probability,Mathematical Physics
A new approach to strong convergence II. The classical ensembles

Chi-Fang Chen,Jorge Garza-Vargas,Ramon van Handel

2024-12-01

Abstract:The first paper in this series introduced a new approach to strong convergence of random matrices that is based primarily on soft arguments. This method was applied to achieve a refined qualitative and quantitative understanding of strong convergence of random permutation matrices and of more general representations of the symmetric group. In this paper, we introduce new ideas that make it possible to achieve stronger quantitative results and that facilitate the application of the method to new models. When applied to the Gaussian ensembles GUE/GOE/GSE of dimension $N$, these methods achieve strong convergence for noncommutative polynomials with matrix coefficients of dimension $\exp(o(N))$. This provides a sharp form of a result of Pisier on strong convergence with coefficients in a subexponential operator space. Analogous results up to logarithmic factors are obtained for Haar-distributed random matrices in $\mathrm{U}(N)/\mathrm{O}(N)/\mathrm{Sp}(N)$. We further illustrate the methods of this paper in three applications: we obtain improved rates for strong convergence of random permutation matrices; we obtain a quantitative form of strong convergence of the model introduced by Hayes for the solution of the Peterson-Thom conjecture; and we prove strong convergence of all nontrivial representations of $\mathrm{SU}(N)$ of dimension up to $\exp(N^{1/3-\delta})$, improving a recent result of Magee and de la Salle.

Probability,Group Theory,Operator Algebras
Eigenvalues and Singular Values of Products of Rectangular Gaussian Random Matrices (The Extended Version)

Z. Burda,A. Jarosz,G. Livan,M. A. Nowak,A. Swiech

DOI: https://doi.org/10.5506/APhysPolB.42.939

2011-03-21

Abstract:We consider a product of an arbitrary number of independent rectangular Gaussian random matrices. We derive the mean densities of its eigenvalues and singular values in the thermodynamic limit, eventually verified numerically. These densities are encoded in the form of the so called M-transforms, for which polynomial equations are found. We exploit the methods of planar diagrammatics, enhanced to the non-Hermitian case, and free random variables, respectively; both are described in the appendices. As particular results of these two main equations, we find the singular behavior of the spectral densities near zero. Moreover, we propose a finite-size form of the spectral density of the product close to the border of its eigenvalues' domain. Also, led by the striking similarity between the two main equations, we put forward a conjecture about a simple relationship between the eigenvalues and singular values of any non-Hermitian random matrix whose spectrum exhibits rotational symmetry around zero.

Statistical Mechanics,Mathematical Physics
Generating random Gaussian states

Leevi Leppäjärvi,Ion Nechita,Ritabrata Sengupta

2024-01-24

Abstract:We develop a method for the random sampling of (multimode) Gaussian states in terms of their covariance matrix, which we refer to as a random quantum covariance matrix (RQCM). We analyze the distribution of marginals and demonstrate that the eigenvalues of an RQCM converge to a shifted semicircular distribution in the limit of a large number of modes. We provide insights into the entanglement of such states based on the positive partial transpose (PPT) criteria. Additionally, we show that the symplectic eigenvalues of an RQCM converge to a probability distribution that can be characterized using free probability. We present numerical estimates for the probability of a RQCM being separable and, if not, its extendibility degree, for various parameter values and mode bipartitions.

Quantum Physics,Mathematical Physics,Probability
Edge Universality for non-Hermitian Random Matrices

Giorgio Cipolloni,László Erdős,Dominik Schröder

DOI: https://doi.org/10.1007/s00440-020-01003-7

2020-09-09

Abstract:We consider large non-Hermitian real or complex random matrices $X$ with independent, identically distributed centred entries. We prove that their local eigenvalue statistics near the spectral edge, the unit circle, coincide with those of the Ginibre ensemble, i.e. when the matrix elements of $X$ are Gaussian. This result is the non-Hermitian counterpart of the universality of the Tracy-Widom distribution at the spectral edges of the Wigner ensemble.

Probability,Mathematical Physics
Optimal Lower Bound on Eigenvector Overlaps for non-Hermitian Random Matrices

Giorgio Cipolloni,László Erdős,Joscha Henheik,Dominik Schröder

2024-01-11

Abstract:We consider large non-Hermitian $N\times N$ matrices with an additive independent, identically distributed (i.i.d.) noise for each matrix elements. We show that already a small noise of variance $1/N$ completely thermalises the bulk singular vectors, in particular they satisfy the strong form of Quantum Unique Ergodicity (QUE) with an optimal speed of convergence. In physics terms, we thus extend the Eigenstate Thermalisation Hypothesis, formulated originally by [Deutsch 1991] and proven for Wigner matrices in [Cipolloni, Erdős, Schröder 2020], to arbitrary non-Hermitian matrices with an i.i.d. noise. As a consequence we obtain an optimal lower bound on the diagonal overlaps of the corresponding non-Hermitian eigenvectors. This quantity, also known as the (square of the) eigenvalue condition number measuring the sensitivity of the eigenvalue to small perturbations, has notoriously escaped rigorous treatment beyond the explicitly computable Ginibre ensemble apart from the very recent upper bounds given in [<a class="link-https" data-arxiv-id="2005.08930" href="https://arxiv.org/abs/2005.08930">arXiv:2005.08930</a>] and [<a class="link-https" data-arxiv-id="2005.08908" href="https://arxiv.org/abs/2005.08908">arXiv:2005.08908</a>]. As a key tool, we develop a new systematic decomposition of general observables in random matrix theory that governs the size of products of resolvents with deterministic matrices in between.

Probability,Mathematical Physics
Wegner estimate and upper bound on the eigenvalue condition number of non-Hermitian random matrices

László Erdős,Hong Chang Ji

2023-06-06

Abstract:We consider $N\times N$ non-Hermitian random matrices of the form $X+A$, where $A$ is a general deterministic matrix and $\sqrt{N}X$ consists of independent entries with zero mean, unit variance, and bounded densities. For this ensemble, we prove (i) a Wegner estimate, i.e. that the local density of eigenvalues is bounded by $N^{1+o(1)}$ and (ii) that the expected condition number of any bulk eigenvalue is bounded by $N^{1+o(1)}$; both results are optimal up to the factor $N^{o(1)}$. The latter result complements the very recent matching lower bound obtained in [15] (<a class="link-https" data-arxiv-id="2301.03549" href="https://arxiv.org/abs/2301.03549">arXiv:2301.03549</a>) and improves the $N$-dependence of the upper bounds in [5,6,32] (<a class="link-https" data-arxiv-id="1906.11819" href="https://arxiv.org/abs/1906.11819">arXiv:1906.11819</a>, <a class="link-https" data-arxiv-id="2005.08930" href="https://arxiv.org/abs/2005.08930">arXiv:2005.08930</a>, <a class="link-https" data-arxiv-id="2005.08908" href="https://arxiv.org/abs/2005.08908">arXiv:2005.08908</a>). Our main ingredient, a near-optimal lower tail estimate for the small singular values of $X+A-z$, is of independent interest.

Probability

A Proletarian Approach to Generating Eigenvalues of GUE Matrices

Distribution of the number of pivots needed using Gaussian elimination with partial pivoting on random matrices

Extreme eigenvalues of Laplacian random matrices with Gaussian entries

PDEs satisfied by extreme eigenvalues distributions of GUE and LUE

Asymptotic Gap Probability Distributions of the Gaussian Unitary Ensembles and Jacobi Unitary Ensembles

Strong convergence for tensor GUE random matrices

The Exact Density of the Eigenvalues of the Wishart and Matrix-Variate Gamma and Beta Random Variables

Computing marginal eigenvalue distributions for the Gaussian and Laguerre orthogonal ensembles

Universal cubic eigenvalue repulsion for random normal matrices

A new approach to the characteristic polynomial of a random unitary matrix

Eigenvalues of random matrices from compact classical groups in Wasserstein metric

On Random Matrices Arising in Deep Neural Networks. Gaussian Case

Asymptotic free independence and entry permutations for Gaussian random matrices. Part II: Infinitesimal freeness

Combinatorial aspects of Parraud's asymptotic expansion for GUE matrices

Edge universality of sparse Erdős-Rényi digraphs

A new approach to strong convergence II. The classical ensembles

Eigenvalues and Singular Values of Products of Rectangular Gaussian Random Matrices (The Extended Version)

Generating random Gaussian states

Edge Universality for non-Hermitian Random Matrices

Optimal Lower Bound on Eigenvector Overlaps for non-Hermitian Random Matrices

Wegner estimate and upper bound on the eigenvalue condition number of non-Hermitian random matrices