Abstract:Algebraic multigrid (AMG) is known to be an effective solver for many sparse symmetric positive definite (SPD) linear systems. For SPD systems, the convergence theory of AMG is well-understood in terms of the $A$-norm, but in a nonsymmetric setting, such an energy norm is non-existent. For this reason, convergence of AMG for nonsymmetric systems of equations remains an open area of research. A particular aspect missing from theory of nonsymmetric and indefinite AMG is the incorporation of general relaxation schemes. In the SPD setting, the classical form of optimal AMG interpolation provides a useful insight in determining the best possible two-grid convergence rate of a method based on an arbitrary symmetrized relaxation scheme. In this work, we discuss a generalization of the optimal AMG convergence theory targeting nonsymmetric problems, using a certain matrix-induced orthogonality of the left and right eigenvectors of a generalized eigenvalue problem relating the system matrix and relaxation operator. We show that using this generalization of the optimal convergence theory, one can obtain a measure of the spectral radius of the two grid error transfer operator that is mathematically equivalent to the derivation in the SPD setting for optimal interpolation, which instead uses norms. In addition, this generalization of the optimal AMG convergence theory can be further extended for symmetric indefinite problems, such as those arising from saddle point systems so that one can obtain a precise convergence rate of the resulting two-grid method based on optimal interpolation. We provide supporting numerical examples of the convergence theory for nonsymmetric advection-diffusion problems, two-dimensional Dirac equation motivated by $\gamma_5$-symmetry, and the mixed Darcy flow problem corresponding to a saddle point system.

New Convergence Analysis of GMRES with Weighted Norms, Preconditioning, and Deflation, Leading to a New Deflation Space

New Convergence Analysis of GMRES with Weighted Norms, Preconditioning and Deflation, Leading to a New Deflation Space

Hermitian Preconditioning for a class of Non-Hermitian Linear Systems

Weighted and deflated global GMRES algorithms for solving large Sylvester matrix equations

Deflated Iterative Methods for Linear Equations with Multiple Right-Hand Sides

Generalized Optimal AMG Convergence Theory for Nonsymmetric and Indefinite Problems

The Convergence of Generalized Lanczos Methods for Large Unsymmetric Eigenproblems

Convergence analysis of inner-iteration preconditioned GMRES

Deflation and Flexible SAP-Preconditioning of GMRES in Lattice QCD Simulation

Convergence Analysis of Gradient Algorithms on Riemannian Manifolds Without Curvature Constraints and Application to Riemannian Mass

Deflation for the Off-Diagonal Block in Symmetric Saddle Point Systems

Preconditioning Low Rank Generalized Minimal Residual Method (GMRES) for Implicit Discretizations of Matrix Differential Equations

Flexible and deflated variants of the block shifted GMRES method

A spectrally preconditioned and initially deflated variant of the restarted block GMRES method for solving multiple right-hand sides linear systems

Deflated GMRES with multigrid for lattice QCD

Block GMRES Method with Inexact Breakdowns and Deflated Restarting

Convergence Properties of Nonlinear Conjugate Gradient Methods.

A Nonlinear GMRES Optimization Algorithm for Canonical Tensor Decomposition

Nonlinear conjugate gradient methods: worst-case convergence rates via computer-assisted analyses

Deflation of eigenvalues for iterative methods in lattice QCD

Randomized flexible GMRES with deflated restarting