A Walk Outside Spheres for the fractional Laplacian: Fields and first eigenvalue

DOI10.1090/MCOM/3422zbMATH Open1416.65034arXiv1803.03921OpenAlexW2964248318WikidataQ128580608 ScholiaQ128580608MaRDI QIDQ5226659FDOQ5226659

Authors: Tony Shardlow 

Publication date: 1 August 2019

Published in: Mathematics of Computation (Search for Journal in Brave)

Abstract: The Feynman-Kac formula for the exterior-value problem for the fractional Laplacian leads to a walk-outside-spheres algorithm via sampling alpha-stable Levy processes on their exit from maximally inscribed balls and sampling their occupation distribution. Kyprianou, Osojnik, and Shardlow (2017) developed this algorithm, providing a complexity analysis and an implementation, for approximating the solution at a single point in the domain. This paper shows how to efficiently sample the whole field by generating an approximation in L_2(D), for a domain D . The method takes advantage of a hierarchy of triangular meshes and uses the multilevel Monte Carlo method for Hilbert space-valued quantities of interest. We derive complexity bounds in terms of the fractional parameter alpha and demonstrate that the method gives accurate results for two problems with exact solutions. Finally, we show how to couple the method with the variable-accuracy Arnoldi iteration to compute the smallest eigenvalue of the fractional Laplacian. A criteria is derived for the variable accuracy and a comparison is given with analytical results of Dyda (2012).

Full work available at URL: https://arxiv.org/abs/1803.03921

Recommendations

• On the Laplacian and fractional Laplacian in an exterior domain
• Eigenvalues of the fractional Laplace operator in the unit ball
• Fractional Laplacians : a short survey
• Eigenvalue bounds for the fractional Laplacian: a review
• On a singular value problem for the fractional Laplacian on the exterior of the unit ball
• Bounds for the eigenvalues of the fractional Laplacian
• Fractional Laplacians on ellipsoids
• On an eigenvalue problem involving the fractional \((s, p)\)-Laplacian
• On fractional Laplacians
• The fractional Laplacian

zbMATH Keywords

eigenvalue problemsfractional LaplacianLevy processesmultilevel Monte CarloArnoldi algorithmwalk on spheresnumerical solution of PDEsexterior-value problems

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Fractional partial differential equations (35R11) Numerical solutions to stochastic differential and integral equations (65C30)

Cites Work

• Title not available (Why is that?)
• Regularity theory for general stable operators
• Numerical methods for large eigenvalue problems
• The Mathematical Theory of Finite Element Methods
• Fractional calculus for power functions and eigenvalues of the fractional Laplacian
• Title not available (Why is that?)
• Quasi-Monte Carlo methods for elliptic PDEs with random coefficients and applications
• The principle of minimized iterations in the solution of the matrix eigenvalue problem
• On the Distribution of First Hits for the Symmetric Stable Processes
• Integration and approximation in arbitrary dimensions
• Inexact inverse iteration for symmetric matrices
• Shift-invert Arnoldi's method with preconditioned iterative solves
• Variable Accuracy of Matrix-Vector Products in Projection Methods for Eigencomputation
• Some observations on the Green function for the ball in the fractional Laplace framework
• Unbiased `walk-on-spheres' Monte Carlo methods for the fractional Laplacian
• Aspects of an adaptive finite element method for the fractional Laplacian: a priori and a posteriori error estimates, efficient implementation and multigrid solver
• Frontiers in numerical analysis. 10th LMS-EPSRC numerical analysis summer school, Durham, UK, July 7--19, 2002
• Towards an efficient finite element method for the integral fractional Laplacian on polygonal domains

Cited In (5)

• Unbiased `walk-on-spheres' Monte Carlo methods for the fractional Laplacian
• A new `walk on spheres' type method for fractional diffusion equation in high dimensions based on the Feynman-Kac formulas
• Efficient Monte Carlo Method for Integral Fractional Laplacian in Multiple Dimensions
• Numerical methods for nonlocal and fractional models
• A deep learning method for computing eigenvalues of the fractional Schrödinger operator

This page was built for publication: A Walk Outside Spheres for the fractional Laplacian: Fields and first eigenvalue

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5226659)