SPEEDING UP KRYLOV SUBSPACE METHODS FOR COMPUTING f(A)b VIA RANDOMIZATION

被引:4
作者
Cortinovis, Alice [1 ]
Kressner, Daniel [2 ]
Nakatsukasa, Yuji [3 ]
机构
[1] Stanford Univ, Dept Math, Stanford, CA 94305 USA
[2] Ecole Polytech Fed Lausanne, Inst Math, CH-1015 Lausanne, Switzerland
[3] Univ Oxford, Math Inst, Oxford OX2 6GG, England
关键词
matrix functions; Krylov subspace method; sketching; nonorthonormal basis; ran- domized algorithms; least-squares problem; MATRIX; ALGORITHM;
D O I
10.1137/22M1543458
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
This work is concerned with the computation of the action of a matrix function f(A), such as the matrix exponential or the matrix square root, on a vector b. For a general matrix A, this can be done by computing the compression of A onto a suitable Krylov subspace. Such compression is usually computed by forming an orthonormal basis of the Krylov subspace using the Arnoldi method. In this work, we propose to compute (nonorthonormal) bases in a faster way and to use a fast randomized algorithm for least -squares problems to compute the compression of A onto the Krylov subspace. We present some numerical examples which show that our algorithms can be faster than the standard Arnoldi method while achieving comparable accuracy.
引用
收藏
页码:619 / 633
页数:15
相关论文
共 26 条
[1]  
Afanasjew M, 2007, ELECTRON T NUMER ANA, V28, P206
[2]   BLENDENPIK: SUPERCHARGING LAPACK'S LEAST-SQUARES SOLVER [J].
Avron, Haim ;
Maymounkov, Petar ;
Toledo, Sivan .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2010, 32 (03) :1217-1236
[3]   RANDOMIZED GRAM-SCHMIDT PROCESS WITH APPLICATION TO GMRES [J].
Balabanov, Oleg ;
Grigori, Laura .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2022, 44 (03) :A1450-A1474
[4]   LOW-RANK UPDATES OF MATRIX FUNCTIONS II: RATIONAL KRYLOV METHODS [J].
Beckermann, Bernhard ;
Cortinovis, Alice ;
Kressner, Daniel ;
Schweitzer, Marcel .
SIAM JOURNAL ON NUMERICAL ANALYSIS, 2021, 59 (03) :1325-1347
[5]  
Benzi Michele, 2020, GAMM - Mitteilungen, V43, DOI 10.1002/gamm.202000012
[6]   Decay Properties of Spectral Projectors with Applications to Electronic Structure [J].
Benzi, Michele ;
Boito, Paola ;
Razouk, Nader .
SIAM REVIEW, 2013, 55 (01) :3-64
[7]   ART: Adaptive residual-time restarting for Krylov subspace matrix exponential evaluations [J].
Botchev, M. A. ;
Knizhnerman, L. A. .
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, 2020, 364
[8]   A Comparison of Accuracy and Efficiency of Parallel Solvers for Fractional Power Diffusion Problems [J].
Ciegis, Raimondas ;
Starikovicius, Vadimas ;
Margenov, Svetozar ;
Kriauziene, Rima .
PARALLEL PROCESSING AND APPLIED MATHEMATICS (PPAM 2017), PT I, 2018, 10777 :79-89
[9]   THE NUMERICAL RANGE IS A (1+√2)-SPECTRAL SET [J].
Crouzeix, M. ;
Palencia, C. .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2017, 38 (02) :649-655
[10]   A Schur-Parlett algorithm for computing matrix functions [J].
Davies, PI ;
Higham, NJ .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2003, 25 (02) :464-485