Abstract:SIAM Journal on Matrix Analysis and Applications, Volume 45, Issue 1, Page 529-552, March 2024. GMRES-based iterative refinement in three precisions (GMRES-IR3), proposed by Carson and Higham in 2018, uses a low precision LU factorization to accelerate the solution of a linear system without compromising numerical stability or robustness. GMRES-IR3 solves the update equation of iterative refinement using GMRES preconditioned by the LU factors, where all operations within GMRES are carried out in the working precision [math], except for the matrix–vector products and the application of the preconditioner, which require the use of extra precision [math]. The use of extra precision can be expensive, and is especially unattractive if it is not available in hardware; for this reason, existing implementations have not used extra precision, despite the absence of an error analysis for this approach. In this article, we propose to relax the requirements on the precisions used within GMRES, allowing the use of arbitrary precisions [math] for applying the preconditioned matrix–vector product and [math] for the rest of the operations. We obtain the five-precision GMRES-based iterative refinement (GMRES-IR5) algorithm which has the potential to solve relatively badly conditioned problems in less time and memory than GMRES-IR3. We develop a rounding error analysis that generalizes that of GMRES-IR3, obtaining conditions under which the forward and backward errors converge to their limiting values. Our analysis makes use of a new result on the backward stability of MGS-GMRES in two precisions. On hardware where three or more arithmetics are available, which is becoming very common, the number of possible combinations of precisions in GMRES-IR5 is extremely large. We provide an analysis of our theoretical results that identifies a relatively small subset of relevant combinations. By choosing from within this subset one can achieve different levels of tradeoff between cost and robustness, which allows for a finer choice of precisions depending on the problem difficulty and the available hardware. We carry out numerical experiments on random dense and SuiteSparse matrices to validate our theoretical analysis and discuss the complexity of GMRES-IR5.

Accelerate Weighted Gmres by Augmenting Error Approximations

Restarted Simpler GMRES Augmented with Harmonic Ritz Vectors and Approximate Errors

Restarted GMRES Method Augmented with the Combination of Harmonic Ritz Vectors and Error Approximations

Solving Shifted Linear Systems with Restarted GMRES Augmented with Error Approximations.

Stochastic Momentum Method with Double Acceleration for Regularized Empirical Risk Minimization

Improving the Performance of the GMRES Method using Mixed-Precision Techniques

Randomized flexible GMRES with deflated restarting

GMRES with randomized sketching and deflated restarting

Adaptive first-order methods with enhanced worst-case rates

Optimized convergence of stochastic gradient descent by weighted averaging

Nesterov's Acceleration For Approximate Newton.

A Note on Nesterov's Accelerated Method in Nonconvex Optimization: a Weak Estimate Sequence Approach

Block GMRES Method with Inexact Breakdowns and Deflated Restarting

Improving GMRES(m) Using an Adaptive Switching Controller

Accelerated Gradient Methods with Gradient Restart: Global Linear Convergence

Convergence Analysis of Accelerated Stochastic Gradient Descent under the Growth Condition

A Proportional-Derivative Control Strategy for Restarting the GMRES(m) Algorithm.

A Preconditioned and Extrapolation-Accelerated GMRES Method for PageRank

A New GMRES(m) Method for Markov Chains

Five-Precision GMRES-Based Iterative Refinement

A Generalized Adaptive Levenberg–Marquardt Method for Solving Nonlinear Ill-Posed Problems