Smallest eigenvalue of large Hankel matrices at critical point: comparing conjecture with parallelised computation

DOI10.1016/J.AMC.2019.124628MaRDI QIDQ2286151zbMATH OpenOpenAlexWikidataFDO

Authors Yang Chen, Jakub Sikorowski, Mengkun Zhu

Publication date 9 January 2020

Published in Applied Mathematics and Computation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1810.01478

zbMATH Keywords

random matrixsmallest eigenvalueextremely ill-conditioned Hankel matricesparallel eigensolver

Mathematics Subject Classification ID

Parallel numerical computation (65Y05) Eigenvalues, singular values, and eigenvectors (15A18) Hermitian, skew-Hermitian, and related matrices (15B57) Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Asymptotic expansions of solutions to ordinary differential equations (34E05) Orthogonal functions and polynomials, general theory of nontrigonometric harmonic analysis (42C05)

Abstract: We propose a novel parallel numerical algorithm for calculating the smallest eigenvalues of highly ill-conditioned matrices. It is based on the {it LDLT} decomposition and involves finding a $kimesk$ sub-matrix of the inverse of the original $NimesN$ Hankel matrix $H_N^{-1}$ . The computation involves extremely high precision arithmetic, message passing interface, and shared memory parallelisation. We demonstrate that this approach achieves good scalability on a high performance computing cluster (HPCC) which constitute a major improvement of the earlier approaches. We use this method to study a family of Hankel matrices generated by the weight $$ supported on $[0,infty)$ and $$ Such weight generates Hankel determinant, a fundamental object in random matrix theory. In the situation where $$ the smallest eigenvalue tend to 0, exponentially fast as $N$ gets large. If $$ the situation where the classical moment problem is indeterminate, the smallest eigenvalue is bounded from below by a positive number for all $N$, including infinity. If $$ it is conjectured that the smallest eigenvalue tends to 0 algebraically, with a precise exponent. The algorithm run on the HPCC producing fantastic match between the theoretical value of $2/pi$ and the numerical result.

Recommendations

• The smallest eigenvalue of large Hankel matrices
• Computing the smallest eigenvalue of large ill-conditioned Hankel matrices
• Small eigenvalues of large Hankel matrices
• The smallest eigenvalue of large Hankel matrices generated by a deformed Laguerre weight
• The smallest eigenvalue of Hankel matrices

Cites work

• scientific article; zbMATH DE number 3134390 (Why is no real title available?)
• scientific article; zbMATH DE number 3561857 (Why is no real title available?)
• scientific article; zbMATH DE number 3219899 (Why is no real title available?)
• scientific article; zbMATH DE number 2228141 (Why is no real title available?)
• Computing the smallest eigenvalue of large ill-conditioned Hankel matrices
• Condition numbers of Hankel matrices for exponential weights
• Estimating the Largest Eigenvalue by the Power and Lanczos Algorithms with a Random Start
• Small Eigenvalues of Large Hankel Matrices
• Small eigenvalues of large Hankel matrices
• Small eigenvalues of large Hankel matrices: The indeterminate case
• Smallest eigenvalues of Hankel matrices for exponential weights
• The condition number of real Vandermonde, Krylov and positive definite Hankel matrices
• The smallest eigenvalue of Hankel matrices
• The smallest eigenvalue of large Hankel matrices
• The smallest eigenvalue of large Hankel matrices generated by a deformed Laguerre weight
• Thermodynamic relations of the Hermitian matrix ensembles

Cited in

(6)

• The smallest eigenvalue of large Hankel matrices generated by a singularly perturbed Laguerre weight
• The smallest eigenvalue of the Hankel matrices associated with a perturbed Jacobi weight
• The smallest eigenvalue of large Hankel matrices associated with a semiclassical Laguerre weight
• Computing the smallest eigenvalue of large ill-conditioned Hankel matrices
• The smallest eigenvalue of large Hankel matrices associated with a singularly perturbed Gaussian weight
• The smallest eigenvalue of large Hankel matrices generated by a deformed Laguerre weight

This page was built for publication: Smallest eigenvalue of large Hankel matrices at critical point: comparing conjecture with parallelised computation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2286151)