DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Modifying AMG Coarse Spaces with Weak Approximation Property to Exhibit Approximation in Energy Norm

Abstract

Algebraic multigrid (AMG) coarse spaces are commonly constructed so that they exhibit the so-called weak approximation property (WAP) which is a necessary and sufficient condition for uniform two-grid convergence. Here, this paper studies a modification of such coarse spaces so that the modified ones provide approximation in energy norm. Our modification is based on the projection in energy norm onto an orthogonal complement of original coarse space. This generally leads to dense modified coarse space matrices, which is hence computationally infeasible. To remedy this, based on the fact that the projection involves inverse of a well-conditioned matrix, we use polynomials to approximate the projection and, therefore, obtain a practical, sparse modified coarse matrix and prove that the modified coarse space maintains computationally feasible approximation in energy norm. We present some numerical results for both PDE discretization matrices as well as graph Laplacian ones, which are in accordance with our theoretical results.

Authors:
 [1];  [2]
  1. Tufts Univ., Medford, MA (United States). Dept. of Mathematics
  2. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States). Center for Applied Scientific Computing; Portland State Univ., OR (United States). Dept. of Mathematics
Publication Date:
Research Org.:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA); National Science Foundation (NSF)
OSTI Identifier:
1669216
Report Number(s):
LLNL-JRNL-744422
Journal ID: ISSN 0895-4798; 899470
Grant/Contract Number:  
AC52-07NA27344; DMS-1619640; DMS-1620063
Resource Type:
Accepted Manuscript
Journal Name:
SIAM Journal on Matrix Analysis and Applications
Additional Journal Information:
Journal Volume: 40; Journal Issue: 3; Journal ID: ISSN 0895-4798
Publisher:
SIAM
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; AMG; strong approximation property; weak approximation property

Citation Formats

Hu, Xiaozhe, and Vassilevski, Panayot S. Modifying AMG Coarse Spaces with Weak Approximation Property to Exhibit Approximation in Energy Norm. United States: N. p., 2019. Web. doi:10.1137/18m1165190.
Hu, Xiaozhe, & Vassilevski, Panayot S. Modifying AMG Coarse Spaces with Weak Approximation Property to Exhibit Approximation in Energy Norm. United States. https://doi.org/10.1137/18m1165190
Hu, Xiaozhe, and Vassilevski, Panayot S. Tue . "Modifying AMG Coarse Spaces with Weak Approximation Property to Exhibit Approximation in Energy Norm". United States. https://doi.org/10.1137/18m1165190. https://www.osti.gov/servlets/purl/1669216.
@article{osti_1669216,
title = {Modifying AMG Coarse Spaces with Weak Approximation Property to Exhibit Approximation in Energy Norm},
author = {Hu, Xiaozhe and Vassilevski, Panayot S.},
abstractNote = {Algebraic multigrid (AMG) coarse spaces are commonly constructed so that they exhibit the so-called weak approximation property (WAP) which is a necessary and sufficient condition for uniform two-grid convergence. Here, this paper studies a modification of such coarse spaces so that the modified ones provide approximation in energy norm. Our modification is based on the projection in energy norm onto an orthogonal complement of original coarse space. This generally leads to dense modified coarse space matrices, which is hence computationally infeasible. To remedy this, based on the fact that the projection involves inverse of a well-conditioned matrix, we use polynomials to approximate the projection and, therefore, obtain a practical, sparse modified coarse matrix and prove that the modified coarse space maintains computationally feasible approximation in energy norm. We present some numerical results for both PDE discretization matrices as well as graph Laplacian ones, which are in accordance with our theoretical results.},
doi = {10.1137/18m1165190},
journal = {SIAM Journal on Matrix Analysis and Applications},
number = 3,
volume = 40,
place = {United States},
year = {Tue Sep 24 00:00:00 EDT 2019},
month = {Tue Sep 24 00:00:00 EDT 2019}
}

Works referenced in this record:

On two-grid convergence estimates
journal, January 2005

  • Falgout, Robert D.; Vassilevski, Panayot S.; Zikatanov, Ludmil T.
  • Numerical Linear Algebra with Applications, Vol. 12, Issue 5-6
  • DOI: 10.1002/nla.437

Localization of Elliptic Multiscale Problems
preprint, January 2011


Abstract robust coarse spaces for systems of PDEs via generalized eigenproblems in the overlaps
journal, August 2013


Lean Algebraic Multigrid (LAMG): Fast Graph Laplacian Linear Solver
journal, January 2012

  • Livne, Oren E.; Brandt, Achi
  • SIAM Journal on Scientific Computing, Vol. 34, Issue 4
  • DOI: 10.1137/110843563

Spectral Upscaling for Graph Laplacian Problems with Application to Reservoir Simulation
journal, January 2017

  • Barker, Andrew T.; Lee, Chak S.; Vassilevski, Panayot S.
  • SIAM Journal on Scientific Computing, Vol. 39, Issue 5
  • DOI: 10.1137/16M1077581

Spectral AMGe ($\rho$AMGe)
journal, January 2003

  • Chartier, T.; Falgout, R. D.; Henson, V. E.
  • SIAM Journal on Scientific Computing, Vol. 25, Issue 1
  • DOI: 10.1137/S106482750139892X

Adaptive Smoothed Aggregation ($\alpha$SA) Multigrid
journal, January 2005

  • Brezina, M.; Falgout, R.; MacLachlan, S.
  • SIAM Review, Vol. 47, Issue 2
  • DOI: 10.1137/050626272

An improved convergence analysis of smoothed aggregation algebraic multigrid: AN IMPROVED ANALYSIS OF SA AMG
journal, March 2011

  • Brezina, Marian; Vaněk, Petr; Vassilevski, Panayot S.
  • Numerical Linear Algebra with Applications, Vol. 19, Issue 3
  • DOI: 10.1002/nla.775

Adaptive AMG with coarsening based on compatible weighted matching
journal, April 2013

  • D’Ambra, Pasqua; Vassilevski, Panayot S.
  • Computing and Visualization in Science, Vol. 16, Issue 2
  • DOI: 10.1007/s00791-014-0224-9

Decay rates for inverses of band matrices
journal, January 1984


On two-grid convergence estimates
journal, January 2005

  • Falgout, Robert D.; Vassilevski, Panayot S.; Zikatanov, Ludmil T.
  • Numerical Linear Algebra with Applications, Vol. 12, Issue 5-6
  • DOI: 10.1002/nla.437

A two-grid SA-AMG convergence bound that improves when increasing the polynomial degree: IMPROVING TG CONVERGENCE WITH INCREASING SMOOTHING STEPS
journal, June 2016

  • Hu, Xiaozhe; Vassilevski, Panayot S.; Xu, Jinchao
  • Numerical Linear Algebra with Applications, Vol. 23, Issue 4
  • DOI: 10.1002/nla.2053

Upscaling of Mixed Finite Element Discretization Problems by the Spectral AMGe Method
journal, January 2016

  • Kalchev, D. Z.; Lee, C. S.; Villa, U.
  • SIAM Journal on Scientific Computing, Vol. 38, Issue 5
  • DOI: 10.1137/15M1036683

Parallel Auxiliary Space AMG for H(Curl) Problems
journal, June 2009

  • Vassilevski, Tzanio V. Kolev Panayot S.
  • Journal of Computational Mathematics, Vol. 27, Issue 5
  • DOI: 10.4208/jcm.2009.27.5.013

Parallel Auxiliary Space AMG Solver for $H(div)$ Problems
journal, January 2012

  • Kolev, Tzanio V.; Vassilevski, Panayot S.
  • SIAM Journal on Scientific Computing, Vol. 34, Issue 6
  • DOI: 10.1137/110859361

Improving the Communication Pattern in Matrix-Vector Operations for Large Scale-Free Graphs by Disaggregation
journal, January 2013

  • Kuhlemann, Verena; Vassilevski, Panayot S.
  • SIAM Journal on Scientific Computing, Vol. 35, Issue 5
  • DOI: 10.1137/12088313X

The Construction of the Coarse de Rham Complexes with Improved Approximation Properties
journal, January 2014

  • Lashuk, Ilya V.; Vassilevski, Panayot S.
  • Computational Methods in Applied Mathematics, Vol. 14, Issue 2
  • DOI: 10.1515/cmam-2014-0004

Energy-minimizing coarse spaces for two-level Schwarz methods for multiscale PDEs
journal, October 2009

  • Van lent, Jan; Scheichl, Robert; Graham, Ivan G.
  • Numerical Linear Algebra with Applications, Vol. 16, Issue 10
  • DOI: 10.1002/nla.641

Lean Algebraic Multigrid (LAMG): Fast Graph Laplacian Linear Solver
journal, January 2012

  • Livne, Oren E.; Brandt, Achi
  • SIAM Journal on Scientific Computing, Vol. 34, Issue 4
  • DOI: 10.1137/110843563

Multilevel upscaling through variational coarsening: MULTILEVEL VARIATIONAL UPSCALING
journal, February 2006

  • MacLachlan, Scott P.; Moulton, J. David
  • Water Resources Research, Vol. 42, Issue 2
  • DOI: 10.1029/2005WR003940

Localization of elliptic multiscale problems
journal, June 2014


The Black Box Multigrid Numerical Homogenization Algorithm
journal, May 1998

  • Moulton, J. David; Dendy, Joel E.; Hyman, James M.
  • Journal of Computational Physics, Vol. 142, Issue 1
  • DOI: 10.1006/jcph.1998.5911

An algebraic multigrid method for finite element discretizations with edge elements: AN ALGEBRAIC MULTIGRID METHOD
journal, February 2002

  • Reitzinger, S.; Schöberl, J.
  • Numerical Linear Algebra with Applications, Vol. 9, Issue 3
  • DOI: 10.1002/nla.271

On Two Ways of Stabilizing the Hierarchical Basis Multilevel Methods
journal, January 1997


Abstract robust coarse spaces for systems of PDEs via generalized eigenproblems in the overlaps
journal, August 2013


Coarse Spaces by Algebraic Multigrid: Multigrid Convergence and Upscaling Error Estimates
journal, April 2011


Algebraic multigrid methods
journal, May 2017