DeepAI AI Chat
Log In Sign Up

Theoretical error estimates for computing the matrix logarithm by Padé-type approximants

by   Lidia Aceto, et al.

In this article, we focus on the error that is committed when computing the matrix logarithm using the Gauss–Legendre quadrature rules. These formulas can be interpreted as Padé approximants of a suitable Gauss hypergeometric function. Empirical observation tells us that the convergence of these quadratures becomes slow when the matrix is not close to the identity matrix, thus suggesting the usage of an inverse scaling and squaring approach for obtaining a matrix with this property. The novelty of this work is the introduction of error estimates that can be used to select a priori both the number of Legendre points needed to obtain a given accuracy and the number of inverse scaling and squaring to be performed. We include some numerical experiments to show the reliability of the estimates introduced.


Fast and accurate approximations to fractional powers of operators

In this paper we consider some rational approximations to the fractional...

A Preconditioning Technique for Computing Functions of Triangular Matrices

We propose a simple preconditioning technique that, if incorporated into...

Afternote to Coupling at a distance: convergence analysis and a priori error estimates

In their article "Coupling at a distance HDG and BEM", Cockburn, Sayas a...

An improved estimate of the inverse binary entropy function

Two estimates for the inverse binary entropy function are derived using ...

Numerical integration rules with improved accuracy close to singularities

Sometimes it is necessary to obtain a numerical integration using only d...

Efficient Scaling and Moving Techniques for Spectral Methods in Unbounded Domains

When using Laguerre and Hermite spectral methods to numerically solve PD...