On Krylov Subspace Approximations to the Matrix Exponential Operator
Open accessThe DOI matches its Crossref record; 29 of the 30 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
Krylov subspace methods for approximating the action of matrix exponentials are analyzed in this paper. We derive error bounds via a functional calculus of Arnoldi and Lanczos methods that reduces the study of Krylov subspace approximations of functions of matrices to that of linear systems of equations. As a side result, we obtain error bounds for Galerkin-type Krylov methods for linear equations, namely, the biconjugate gradient method and the full orthogonalization method. For Krylov approximations to matrix exponentials, we show superlinear error decay from relatively small iteration numbers onwards, depending on the geometry of the numerical range, the spectrum, or the pseudospectrum. The convergence to exp$(\tau A)v$ is faster than that of corresponding Krylov methods for the solution of linear equations $(I-\tau A)x=v$, which usually arise in the numerical solution of stiff ordinary differential equations (ODEs). We therefore propose a new class of time integration methods for large systems of nonlinear differential equations which use Krylov approximations to the exponential function of the Jacobian instead of solving linear or nonlinear systems of equations in every time step.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Analogous to the eigenvalue problem in DMRG, both Eqs. ( 7) and ( 8) can be integrated using a Lanczos scheme [28] in order to keep the computational cost at O(D 3 ).”
From Unifying time evolution and optimization with matrix product states (2016) · cites this paper as [28]
Citations
Cites
Solving Ordinary Differential Equations II
Hairer, E., & Wanner, G. (1996) · Springer Series in Computational Mathematics · doi:10.1007/978-3-642-05221-7 · Cited by 4,342
Solving Ordinary Differential Equations II
Hairer, E., & Wanner, G. (1991) · Springer Series in Computational Mathematics · doi:10.1007/978-3-662-09947-6 · Cited by 1,456
Theory of Functions of a Complex Variable
Markushevich, A. I., Silverman, R. A., & Gillis, J. (1966) · Physics Today · doi:10.1063/1.3048359 · Cited by 1,211
Unitary quantum time evolution by iterative Lanczos reduction
Park, T. J., & Light, J. C. (1986) · The Journal of Chemical Physics · doi:10.1063/1.451548 · Cited by 888
Uniform Numerical Methods for Problems with Initial and Boundary Layers.
Berger, A. E., Doolan, E. P., Miller, J. J. H., & Schilders, W. H. A. (1982) · Mathematics of Computation · doi:10.2307/2007353 · Cited by 754
Analysis of Some Krylov Subspace Approximations to the Matrix Exponential Operator
Saad, Y. (1992) · SIAM Journal on Numerical Analysis · doi:10.1137/0729014 · Cited by 670
Numerical Ranges of Operators on Normed Spaces and of Elements of Normed Algebras
Bonsall, F. F., & Duncan, J. (1971) · Cambridge University Press eBooks · doi:10.1017/cbo9781107359895 · Cited by 614
Propagation Methods for Quantum Molecular Dynamics
Kosloff, R. (1994) · Annual Review of Physical Chemistry · doi:10.1146/annurev.pc.45.100194.001045 · Cited by 602
Krylov subspace methods for solving large unsymmetric linear systems
Saad, Y. (1981) · Mathematics of Computation · doi:10.1090/s0025-5718-1981-0616364-6 · Cited by 449
Generalized Runge-Kutta Processes for Stable Systems with Large Lipschitz Constants
Lawson, J. D. (1967) · SIAM Journal on Numerical Analysis · doi:10.1137/0704033 · Cited by 328
Research works printed but not linked: 5
- A. Friedli, Verallgemeinerte Runge‐Kutta Verfahren zur Lösung steifer Differentialgleichungen , in Lecture Notes in Mathematics 631, Springer‐Verlag, Berlin, 1978, pp. 35–50. · doi:10.1007/bfb0067462
- Hochbruck M., 1997, SIAM J. Sci. Comput., 18
- Knizhnerman L. A., 1991, Comput. Math. and Math. Phys., 31, 1
- Knizhnerman L. A., 1992, Comput. Math. and Math. Phys., 32, 1199
- D. R. Taylor, Analysis of the look‐ahead Lanczos algorithm , Ph.D. thesis, University of California at Berkeley, Berkeley, CA, 1982.
Cited by
Items 301–400 of 810
Exponential Taylor methods: Analysis and implementation
Koskela, A., & Ostermann, A. (2012) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2012.06.004 · Cited by 21
“The convergence of the Krylov approximation (32) generally depends on the field of values of the matrix [11].”
BackgroundChebyshev and conjugate gradient filters for graph image denoising
Tian, D., Mansour, H., Knyazev, A., & Vetro, A. (2014) · IEEE International Conference on Multimedia and Expo workshops · doi:10.1109/icmew.2014.6890711 · Cited by 21
On RD‐rational Krylov approximations to the core‐functions of exponential integrators
Moret, I. (2007) · Numerical Linear Algebra with Applications · doi:10.1002/nla.532 · Cited by 21
“[5, 6]), the convergence does not depend on the norm of A.”
BackgroundStability control for approximate implicit time stepping schemes with minimum residual iterations
Botchev, M. A., Sleijpen, G. L. G., & van der Vorst, H. A. (1999) · Applied Numerical Mathematics · doi:10.1016/s0168-9274(98)00138-x · Cited by 21
“Another approach, that is based on Krylov subspace information, is to use an approximation for the vector e∆tAyn (where A is the Jacobian), for accurate time integration of stiff systems [8, 9].”
Probabilistic relaxation labelling using the Fokker–Planck equation
Wang, H., & Hancock, E. R. (2008) · Pattern Recognition · doi:10.1016/j.patcog.2008.03.030 · Cited by 23
“[56] the values of e−tFv are computed directly by approximating the product of the exponential of the matrix and the vector by finding an element of the Krylov subspace Km = {v0, (tF)v0, .”
MethodZaslavsky, M., & Druskin, V. L. (2010) · Journal of Computational Physics · doi:10.1016/j.jcp.2010.03.019 · Cited by 20
“Krylov subspace methods (originally intended for spectral problems and linear algebraic systems) became an increasedly popular tool for the solution of parabolic equations (the matrix exponential function) in 1980s [14] [16] [21] [28] [32].”
BackgroundLivermore, P. W. (2006) · Journal of Computational Physics · doi:10.1016/j.jcp.2006.05.029 · Cited by 21
“However, even in cases where this discretised matrix is large and dense, various procedures exist to speed up the necessary computations, including Schur factorisation [10] or Krylov subspace methods [11].”
Bhatt, H., & Chowdhury, A. (2020) · International Journal of Computer Mathematics · doi:10.1080/00207160.2020.1814262 · Cited by 20
“Some of the well-known techniques include (i) the rational approximation of φμ(kL) [8,9], (ii) the Krylov subspacemethod [17,41], and (iii) the polynomial approximation of φμ(kL) [10,40].”
MethodFlesch, A., Gorelov, E., Koch, E., & Pavarini, E. (2013) · Physical Review B · doi:10.1103/physrevb.87.195141 · Cited by 20
Turner, I., Ilić, M., & Perré, P. (2011) · Drying Technology · doi:10.1080/07373937.2011.596300 · Cited by 20
“For both formulations given above we step the ODE system in time from its initial state ûð0Þ using the second order exponential Euler integrator scheme (EEM)([24])”
MethodInfluentialSolving linear initial value problems by Faber polynomials
Novati, P. (2002) · Numerical Linear Algebra with Applications · doi:10.1002/nla.287 · Cited by 20
Magnus integrators on multicore CPUs and GPUs
Auer, N., Einkemmer, L. D., Kandolf, P., & Ostermann, A. (2018) · Computer Physics Communications · doi:10.1016/j.cpc.2018.02.019 · Cited by 20
“ne using fast Fourier techniques. However, for many other interesting problems this is not possible. For the latter case a number of approaches have been proposed in the literature (see, for example, [4, 10, 14, 18, 19, 21]). Most of them are based on polynomial interpolation. In [10, 11] it was shown that interpolati…”
MethodModel predictive control of piezo-actuated structures using reduced order models
Rosenzweig, P., Kater, A., & Meurer, T. (2018) · Control Engineering Practice · doi:10.1016/j.conengprac.2018.08.001 · Cited by 20
Jawecki, T., Auzinger, W., & Koch, O. (2019) · BIT Numerical Mathematics · doi:10.1007/s10543-019-00771-6 · Cited by 24
An error analysis of the modified scaling and squaring method
Koikari, S. (2007) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2006.04.032 · Cited by 20
The Jacobi stochastic volatility model
Ackerer, D., Filipović, D., & Pulido, S. (2018) · Finance and Stochastics · doi:10.1007/s00780-018-0364-8 · Cited by 19
“Note that specific numerical methods have been developed to compute the action of the matrix exponential eGT on the basis vn; see, for example, Al-Mohy and Higham [5], Hochbruck and Lubich [36] and references therein.”
A matrix approach for partial differential equations with Riesz space fractional derivatives
Popolizio, M. (2013) · The European Physical Journal Special Topics · doi:10.1140/epjst/e2013-01978-8 · Cited by 19
Using Nonequilibrium Dynamics to Probe Competing Orders in a Mott-Peierls System
Wang, Y., Moritz, B., Chen, C.-C., Jia, C., van Veenendaal, M., & Devereaux, T. (2016) · Physical Review Letters · doi:10.1103/physrevlett.116.086401 · Cited by 19
Error bounds and estimates for Krylov subspace approximations of Stieltjes matrix functions
Frommer, A., & Schweitzer, M. (2015) · BIT Numerical Mathematics · doi:10.1007/s10543-015-0596-3 · Cited by 19
“, for the exponential function [30], for the sign function [42] or for rational functions [18, 20, 21].”
BackgroundMethodEPIRK-W and EPIRK-K Time Discretization Methods
Narayanamurthi, M., Tranquilli, P., Sandu, A., & Tokman, M. (2018) · Journal of Scientific Computing · doi:10.1007/s10915-018-0761-3 · Cited by 21
“ds to a number of trees that arise in ROW order conditions theory to become isomorphic to one another, leading to a dramatic reduction in the total number of order conditions. Exponential integrators [8, 10, 17, 20, 21] are a class of timestepping methods that replace the linear system solves in Rosenbrock style metho…”
MethodWang, H., Xu, L., Li, B., Descombes, S., & Lanteri, S. (2017) · IEEE Transactions on Antennas and Propagation · doi:10.1109/tap.2017.2752223 · Cited by 21
“The second kind of methods include Krylov subspace methods [23]-[36][22] and Chebyshev polynomials based methods [37].”
MethodLimited‐memory polynomial methods for large‐scale matrix functions
Güttel, S., Kreßner, D., & Lund, K. (2020) · GAMM-Mitteilungen · doi:10.1002/gamm.202000019 · Cited by 24
“tization of the 4 (unbounded) Laplace operator on , and as such it is symmetric, large, and sparse. It also tends to have a wide spectrum, which hampers the convergence of (polynomial) Krylov methods [64] for the matrix exponential, while rational Krylov methods are immune to this problem [53]. The solution of (shifte…”
MethodInfluentialÁlvarez, G., da Silva, L. G. G. V. D., Ponce, E., & Dagotto, E. R. (2011) · Physical Review E · doi:10.1103/physreve.84.056706 · Cited by 24
Exponential time-differencing with embedded Runge–Kutta adaptive step control
Whalen, P., Brio, M., & Moloney, J. V. (2014) · Journal of Computational Physics · doi:10.1016/j.jcp.2014.09.038 · Cited by 25
“For large scale non-diagonal systems, several efficient exponential integration methods [15,16] have been designed using Krylov subspace methods [17,18].”
MethodMukam, J. D., & Tambue, A. (2017) · Journal of Scientific Computing · doi:10.1007/s10915-017-0475-y · Cited by 19
“The exponential matrix function φ1 is computed by Krylov subspace technique with fixed dimension m = 10 and tolerance tol = 10−6 [9,31,33].”
MethodBrenes, M., Varma, V. K., Scardicchio, A., & Girotto, I. (2018) · Computer Physics Communications · doi:10.1016/j.cpc.2018.08.010 · Cited by 21
Iserles, A., Kropielnicka, K., & Singh, P. (2018) · SIAM Journal on Numerical Analysis · doi:10.1137/17m1149833 · Cited by 19
“…highly promising superlinear decay of error in the case of the Lanczos method for approximating the matrix exponential is not seen until the number of iterations is larger than the spectral radius of Θ m ( h ) (Hochbruck & Lubich 1997), which is very large unless the time step h is suitably small.”
MethodPreconditioning the Matrix Exponential Operator with Applications
Castillo, P. E., & Saad, Y. (1998) · Journal of Scientific Computing · doi:10.1023/a:1023219016301 · Cited by 19
Practical Implementation of Krylov Subspace Spectral Methods
Lambers, J. V. (2007) · Journal of Scientific Computing · doi:10.1007/s10915-007-9140-1 · Cited by 19
“The preceding result can be compared to the accuracy achieved by an algorithm described by Hochbruck and Lubich in [13] for computing e v for a given matrix A and vector v using the unsymmetric Lanczos algorithm.”
MethodInfluentialMarkovian embedding procedures for non-Markovian stochastic Schrödinger equations
Li, X. (2020) · Physics Letters A · doi:10.1016/j.physleta.2020.127036 · Cited by 19
Graph-Based Rate Control in Pathology Imaging With Lossless Region of Interest Coding
Sánchez, V., & Hernández-Cabronero, M. (2018) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2018.2824819 · Cited by 18
“We take advantage of this fact to reduce the computational complexity by using the Krylov subspace projection technique [51], which is an iterative method for sparse matrix problems.”
MethodA semi-analytical approach to molecular dynamics
Michels, D. L., & Desbrun, M. (2015) · Journal of Computational Physics · doi:10.1016/j.jcp.2015.10.009 · Cited by 19
“To determine a conservative value of m such as the Krylov-based evaluations of matrix functions remain below a given error threshold, we follow [27]: the absolute error”
BackgroundMethodInexact Arnoldi residual estimates and decay properties for functions of non-Hermitian matrices
Pozza, S., & Simoncini, V. (2019) · BIT Numerical Mathematics · doi:10.1007/s10543-019-00763-6 · Cited by 18
Iyengar, G., Phillips, D. J., & Stein, C. (2005) · Lecture notes in computer science · doi:10.1007/11496915_12 · Cited by 18
Carbonell, F., Jiménez, J. C., & Biscay, R. (2002) · Applied Mathematics and Computation · doi:10.1016/s0096-3003(01)00123-0 · Cited by 20
“The LL method [6–10], also called exponentially fitted Euler method [11,12] or Euler exponential method [13,14], is a numerical method for the integration of ODEs.”
MethodA Bunch of Time Integrators for Quantum/Classical Molecular Dynamics
Hochbruck, M., & Lubich, C. (1999) · Lecture notes in computational science and engineering · doi:10.1007/978-3-642-58360-5_24 · Cited by 18
“More recently, the excellent convergence properties of this approach have been clarified in [7,12].”
BackgroundSelf-citationOn convergence of Krylov subspace approximations of time-invariant self-adjoint dynamical systems
Druskin, V. L., & Zaslavsky, M. (2011) · Linear Algebra and its Applications · doi:10.1016/j.laa.2011.02.039 · Cited by 20
“The error of the Krylov or rational Krylov subspace approximations of a function of a non-Hermitian matrix can be estimated via polynomial or rational approximation error on its numerical range as respectively in [20, 22] or [4] obtained via the Faber transform.”
BackgroundMethodInfluentialExponential time differencing for mimetic multilayer ocean models
Pieper, K., Sockwell, K., & Gunzburger, M. (2019) · Journal of Computational Physics · doi:10.1016/j.jcp.2019.108900 · Cited by 17
“For instance, in the skew-symmetric case [17, Theorem 4], after a minimum of M ≥ ∆t | A | iterations, where | A | is the spectral radius of A , the error decreases at an exponential rate.”
BackgroundMeng, X., Hoang, T.-T.-P., Wang, Z., & Ju, L. (2020) · Communications in Computational Physics · doi:10.4208/cicp.oa-2019-0214 · Cited by 17
Down-conversion processes inab initiononrelativistic quantum electrodynamics
Welakuh, D. M., Ruggenthaler, M., Djouom, M.-L. M. T., Appel, H., & Rubio, Á. (2021) · Physical Review Research · doi:10.1103/physrevresearch.3.033067 · Cited by 17
“We solve the time-dependent Schrödinger equation of the coupled light-matter system with a Lanczos propagation scheme [50].”
BackgroundMethodA Class of Exponential Integrators Based on Spectral Deferred Correction
Buvoli, T. (2020) · SIAM Journal on Scientific Computing · doi:10.1137/19m1256166 · Cited by 17
“ve been compared extensively [1]. We brie y outline a method based on scaling and squaring [24] and a method based on contour integration [22]. Other approaches involve Krylov subspace approximations [17, 16] and improved contour integrals [41] but we do not consider them in this paper. 13 4.2.1 Taylor/Pade Scaling an…”
MethodInfluentialLambers, J. V. (2008) · Cited by 17
Ultrafast intersystem crossings in Fe-Co Prussian blue analogues
van Veenendaal, M. (2017) · Scientific Reports · doi:10.1038/s41598-017-06664-4 · Cited by 20
Liang, H., Xiao, X.-R., Gong, Q., & Peng, L.-Y. (2017) · Journal of Physics B Atomic Molecular and Optical Physics · doi:10.1088/1361-6455/aa82d0 · Cited by 17
Exponential collocation methods for conservative or dissipative systems
Wang, B., & Wu, X. (2019) · Journal of Computational and Applied Mathematics · doi:10.1016/j.cam.2019.04.015 · Cited by 17
“It follows form (13) that Here the ¯ ϕ -functions (see, e.g. [30, 32, 33, 34]) are defined by: It is noted that a number of approaches have been developed which work with the application of the ϕ -functions on a vector (see [1, 28, 33], for example).”
Jian, H.-Y., Huang, T., Gu, X., Zhao, X.-L., & Zhao, Y.-L. (2020) · Journal of Computational and Applied Mathematics · doi:10.1016/j.cam.2020.112935 · Cited by 17
Residual and Restarting in Krylov Subspace Evaluation of the $\varphi$ Function
Botchev, M. A., Knizhnerman, L., & Tyrtyshnikov, E. E. (2021) · SIAM Journal on Scientific Computing · doi:10.1137/20m1375383 · Cited by 20
Local Linearization Method for Numerical Integration of Delay Differential Equations
Jiménez, J. C., Pedroso, L., Carbonell, F., & Hernández, V. (2006) · SIAM Journal on Numerical Analysis · doi:10.1137/040607356 · Cited by 17
“, those based on rational Padé approximations [18], the Schur decomposition [18] or Krylov subspace methods [22] (for a recent review see [38]).”
MethodFaragó, I., Horváth, R., & Schilders, W. H. A. (2005) · International Journal of Numerical Modelling Electronic Networks Devices and Fields · doi:10.1002/jnm.570 · Cited by 18
The Gautschi time stepping scheme for edge finite element discretizations of the Maxwell equations
Botchev, M. A., Harutyunyan, D., & van der Vegt, J. J. W. (2006) · Journal of Computational Physics · doi:10.1016/j.jcp.2006.01.014 · Cited by 17
“Algorithms for this have been developed and used in different contexts (we list in the chronological order [37,6,21,32,7,16,8,18], see also Chapter 11 in the recent book [38]).”
BackgroundMethodAdvanced optimal control methods for spin systems
Goodwin, D. L. (2017) · ePrints Soton (University of Southampton) · doi:10.5258/soton/t0003 · Cited by 28
“This type of problem is readily solved with Krylov subspace methods, such as the Arnoldi process [9] or the Lanczos process [173], avoiding explicit calculation of the matrix exponential [106,127,128, 256].”
MethodA New Class of High-Order Methods for Multirate Differential Equations
Luân, V. T., Chinomona, R., & Reynolds, D. R. (2020) · SIAM Journal on Scientific Computing · doi:10.1137/19m125621x · Cited by 16
“now of no previous methods having order ve or higher. Among numerical methods that use the same time step for all components of (1.1), exponential integrators have shown great promise in recent years [6,18,19, 20,21,22,30,31,32,34,28,35]. Most such methods require the approximation of products of matrix functions with…”
ResultInfluentialLai, C.-Y., & Chien, C.-C. (2016) · Scientific Reports · doi:10.1038/srep37256 · Cited by 18
Wang, Y., Devereaux, T., & Chen, C.-C. (2018) · Physical review. B./Physical review. B · doi:10.1103/physrevb.98.245106 · Cited by 16
Kandolf, P., Koskela, A., Relton, S. D., & Schweitzer, M. (2021) · Numerical Linear Algebra with Applications · doi:10.1002/nla.2401 · Cited by 19
“orefocusontwo classes offunctionswhicharise frequentlyin applications.Thefirst class consists of analytic functions represented via the Cauchy integral formula, i.e., f(A) = 1 2ˇi Ê Γ f(t)(tI −A)−1 dt (2) where Γ is a path in the complex plane that winds around spec(A), the spectrum of A, exactly once. The most promine…”
BackgroundMethodInfluentialUga, K. C., Min, M., Lee, T., & Fischer, P. (2012) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2011.12.059 · Cited by 18
Explicit Time Integration Scheme Using Krylov Subspace Method for Reactor Kinetics Equation
Ban, Y., Endo, T., Yamamoto, A., & Yamane, Y. (2011) · Journal of Nuclear Science and Technology · doi:10.1080/18811248.2011.9711698 · Cited by 17
Interpolating functions of matrices on zeros of quasi-kernel polynomials
Moret, I., & Novati, P. (2004) · Numerical Linear Algebra with Applications · doi:10.1002/nla.393 · Cited by 16
KRYLOV SUBSPACE SPECTRAL METHODS FOR VARIABLE-COEFFICIENT INITIAL-BOUNDARY VALUE PROBLEMS
Lambers, J. V. (2005) · Cited by 16
Effective many-body parameters for atoms in nonseparable Gaussian optical potentials
Wall, M. L., Hazzard, K. R. A., & Rey, A. M. (2015) · Physical Review A · doi:10.1103/physreva.92.013610 · Cited by 19
Stochastic exponential integrators for a finite element discretisation of SPDEs with additive noise
Lord, G. J., & Tambue, A. (2018) · Applied Numerical Mathematics · doi:10.1016/j.apnum.2018.10.008 · Cited by 17
A Survey on Methods for Computing Matrix Exponentials in Numerical Schemes for ODEs
Buono, N. D., & Lopez, L. (2003) · Lecture notes in computer science · doi:10.1007/3-540-44862-4_13 · Cited by 15
Numerical solution of conservative finite-dimensional stochastic Schrödinger equations
Mora, C. M. (2005) · The Annals of Applied Probability · doi:10.1214/105051605000000403 · Cited by 15
“Furthermore, Hochbruck and Lubich [12] proved that the convergence of Krylov methods for exp(Jb(Tm, Vm)T/M)u is faster than that for the solution of the linear equation (I−Jb(Tm, Vm)T/M)x= u, which is required in both methods V 1 and the usual implicit Euler scheme when b is linear.”
InfluentialSimplified formulas for the mean and variance of linear stochastic differential equations
Jiménez, J. C. (2015) · Applied Mathematics Letters · doi:10.1016/j.aml.2015.04.009 · Cited by 15
“But, in addition, the new formulas allow the efficient computation of the mean and variance of high dimensional linear SDE (1) by means of the Krylov subspace method [9, 19] for exponential matrices, which is crucial in many practical”
Preconditioned implicit-exponential integrators (IMEXP) for stiff PDEs
Luân, V. T., Tokman, M., & Rainwater, G. (2017) · Journal of Computational Physics · doi:10.1016/j.jcp.2017.01.054 · Cited by 15
A Krylov Subspace Method for the Approximation of Bivariate Matrix Functions
Kreßner, D. (2019) · Springer INdAM series · doi:10.1007/978-3-030-04088-8_10 · Cited by 15
Numerical Integrators for Quantum-Classical Molecular Dynamics
Nettesheim, P., & Schütte, C. (1999) · Lecture notes in computational science and engineering · doi:10.1007/978-3-642-58360-5_22 · Cited by 15
Quadrature Rules Based on the Arnoldi Process
Daniela Calvetti, A., Kim, S., & Reichel, L. (2005) · SIAM Journal on Matrix Analysis and Applications · doi:10.1137/s0895479803423822 · Cited by 15
“, [12] for discussions of related representations.”
Fast Exponential Time Integration for Pricing Options in Stochastic Volatility Jump Diffusion Models
Pang, H., & Sun, H. (2014) · East Asian Journal on Applied Mathematics · doi:10.4208/eajam.280313.061013a · Cited by 15
“However, theoretical analyses [17, 30] show that the”
MethodFast sampling from a Gaussian Markov random field using Krylov subspace approaches
Simpson, D., Turner, I., & Pettitt, A. N. (2008) · QUT ePrints (Queensland University of Technology) · Cited by 20
Adaptive optimal control of entangled qubits
Goodwin, D. L., Singh, P., & Foroozandeh, M. (2022) · Science Advances · doi:10.1126/sciadv.abq4244 · Cited by 14
Error Bounds for Lanczos-Based Matrix Function Approximation
Chen, T., Greenbaum, A., Musco, C., & Musco, C. (2022) · SIAM Journal on Matrix Analysis and Applications · doi:10.1137/21m1427784 · Cited by 14
“Another CIF based approach is [30] which shows that Lanczos-FA exhibits superlinear convergence for the matrix exponential and certain other specific analytic functions.”
BackgroundParallel density matrix propagation in spin dynamics simulations
Edwards, L. J., & Kuprov, I. (2012) · The Journal of Chemical Physics · doi:10.1063/1.3679656 · Cited by 14
Locally exact modifications of numerical schemes
Cieśliński, J. L. (2013) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2013.04.015 · Cited by 14
“Bf MSC 2000: 65P10; 65L12; 34K28 Key words and phrases: geometric numerical integration, exact discretization, locally exact methods, linearization-preserving integrators, exponential integrators, discrete gradient method, Hamiltonian systems, linear stability ∗e-mail: janek@ alpha.uwb.edu.pl”
MethodMinimum action transition paths connecting minima on an energy surface
Koehl, P. (2016) · The Journal of Chemical Physics · doi:10.1063/1.4966974 · Cited by 14
Iserles, A., Kropielnicka, K., & Singh, P. (2018) · Journal of Computational Physics · doi:10.1016/j.jcp.2018.09.047 · Cited by 14
Exponential integrators for the incompressible Navier-Stokes equations.
Newman, C. K. (2004) · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · doi:10.2172/975250 · Cited by 14
“The following results from [41] specify the error bounds in Theorem 2.”
InfluentialSchmidt, M. (2007) · DepositOnce · doi:10.14279/depositonce-1600 · Cited by 14
“considered in [159, 214, 213, 224].”
The Radau--Lanczos Method for Matrix Functions
Frommer, A., Lund, K., Schweitzer, M., & Szyld, D. B. (2017) · SIAM Journal on Matrix Analysis and Applications · doi:10.1137/16m1072565 · Cited by 13
“Applications requiring f(A)b include the solution of differential equations via exponential integrators [18, 19, 20]; determining the communicability—or other quantities of interest—in network analysis [2, 8]; and simulations involving the Neuberger overlap operator or the rational hybrid Monte Carlo algorithm in latt…”
MethodExponential integrators for stochastic Schrödinger equations
Li, J., & Li, X. (2020) · Physical review. E · doi:10.1103/physreve.101.013312 · Cited by 13
3D large-scale transient electromagnetic modeling using a Shift-and-Invert Krylov subspace method
Lu, K., Zhou, J., Li, X., Fan, Y., Qi, Z., & Cao, H. (2022) · Journal of Applied Geophysics · doi:10.1016/j.jappgeo.2022.104573 · Cited by 13
Caliari, M., Cassini, F., Einkemmer, L. D., & Ostermann, A. (2024) · SIAM Journal on Scientific Computing · doi:10.1137/23m1562056 · Cited by 13
“For that purpose Krylov subspace methods [21], schemes based on Leja interpolation [8, 9], and Taylor methods [1] are commonly used.”
MethodJiménez, J. C., & Carbonell, F. (2014) · Journal of Computational and Applied Mathematics · doi:10.1016/j.cam.2014.10.021 · Cited by 14
Approximation of Semigroups and Related Operator Functions by Resolvent Series
Grimm, V., & Gugat, M. (2010) · SIAM Journal on Numerical Analysis · doi:10.1137/090768084 · Cited by 13
“The exp4 integrator internally uses the Krylov subspace method proposed in [11] for the approximation of the φ-functions.”
BackgroundMethodInfluentialImplementation of Parallel Adaptive-Krylov Exponential Solvers for Stiff Problems
Loffeld, J. J., & Tokman, M. (2014) · SIAM Journal on Scientific Computing · doi:10.1137/13094462x · Cited by 16
A posteriori error estimates of krylov subspace approximations to matrix functions
Jia, Z., & Lv, H. (2014) · Numerical Algorithms · doi:10.1007/s11075-014-9878-0 · Cited by 16
“The results in [33] also reveal the superlinear convergence of the approximations, but they are less sharp than those in [9, 17, 36], at least experimentally [21].”
On the Numerical Solution of Nonlinear Schr¨ odinger Type Equations in Fiber Optics
Hohage, T., & Schmidt, F. (2002) · Cited by 13
Fast Time Integration of Navier-Stokes Equations with an Exponential-Integrator Scheme
Li, S., Wang, Z. J., Ju, L., & Luo, L. (2018) · AIAA Aerospace Sciences Meeting · doi:10.2514/6.2018-0369 · Cited by 13
Bhatt, H., Khaliq, A. Q. M., & Wade, B. A. (2018) · Applied Mathematics and Computation · doi:10.1016/j.amc.2018.06.025 · Cited by 16
3-D Large-Scale TEM Modeling Using Restarting Polynomial Krylov Method
Zhou, J., Lu, K., Li, X., Liu, W., Qi, Z., & Qi, Y. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3126291 · Cited by 15
“The memory usage of the polynomial Krylov method mainly includes two parts: the storage of the spatially discrete coefficient matrix and the basis matrix of the Krylov subspace [38], [39].”
MethodA quadrature based method for evaluating exponential-type functions for exponential methods
López-Fernández, M. (2010) · BIT Numerical Mathematics · doi:10.1007/s10543-010-0273-5 · Cited by 13
“where Vm contains the Arnoldi or Lanczos basis of the mth Krylov subspace with respect to A, v and τ , Hm is a much smaller matrix than A, and e1 is the m-dimensional unit vector [ 6 , 7]. We can apply (1.9) to evaluate the products φ(τH m)e1.”
BackgroundInfluentialLambers, J. V. (2009) · Numerical Algorithms · doi:10.1007/s11075-009-9278-z · Cited by 13
Breakdown of the nonrelativistic approximation in superintense laser-matter interactions
Førre, M. (2019) · Physical Review A · doi:10.1103/physreva.99.053410 · Cited by 13
Quantum mechanical MRI simulations: Solving the matrix dimension problem
Allami, A. J., Concilio, M. G., Lally, P., & Kuprov, I. (2019) · Science Advances · doi:10.1126/sciadv.aaw8962 · Cited by 15
Derivation of high-order spectral methods for time-dependent PDE using modified moments.
Lambers, J. V. (2007) · Cited by 13
A Krylov Subspace Method for Option PricingPreprint
Niesen, J., & Wright, W. M. (2011) · SSRN Electronic Journal · doi:10.2139/ssrn.1799124 · Cited by 15
“The most popular ADI methods were developed by Craig and Sneyd (1988) and by Hundsdorfer and Verwer (2003); their popularity stems from the fact that they can handle PDEs with mixed spatial derivatives explicitly and the remaining derivatives separately, but implicitly, while maintaining overall unconditional stabilit…”
InfluentialVinding, M. S., Goodwin, D. L., Kuprov, I., & Lund, T. E. (2021) · Journal of Magnetic Resonance · doi:10.1016/j.jmr.2021.107094 · Cited by 14
“The midpoint variant was inspired by the central finite-difference gradient, although it has a deep theoretical basis in the exponential midpoint method [45– 47, The midpoint method for this manuscript is r J ð MID Þ which is an average of the ð n (cid”
Quantum reservoir complexity by the Krylov evolution approach
Domingo, L., Borondo, F., Scialchi, G. F., Roncaglia, A. J., Carlo, G. G., & Wisniacki, D. A. (2024) · Physical Review A · doi:10.1103/physreva.110.022446 · Cited by 16
Exponential Time Differencing Gauge Method for Incompressible Viscous Flows
Ju, L., & Wang, Z. (2017) · Communications in Computational Physics · doi:10.4208/cicp.oa-2016-0234 · Cited by 12
Meshfree Exponential Integrators
Caliari, M., Ostermann, A., & Rainer, S. (2013) · SIAM Journal on Scientific Computing · doi:10.1137/100818236 · Cited by 12
“, [4, 13, 15].”
Metadata
- Authors
- Hochbruck, M., & Lubich, C.
- Affiliations
- University of Tübingen (DE)
- Venue
- SIAM Journal on Numerical Analysis
- Published
- 1997-10
- Volume
- 34
- Issue
- 5
- Pages
- 1911-1925
- ISSN
- 0036-1429 (ISSN-L)
- 1095-7170
- OpenAlex
- W1984347942
- Cited by
- 795
- References
- 30
- Citations by year
- 2026: 27 · 2025: 23 · 2024: 19 · 2023: 23 · 2022: 23 · 2021: 37 · 2020: 42 · 2019: 41 · 2018: 46 · 2017: 41 · 2016: 42 · 2015: 32 · 2014: 44 · 2013: 44 · 2012: 30
- FWCI
- 14.18
- Type
- journal article
- Language
- en
- Open access
- green
- doi.org
- Licence
- other-oa
- Locations
- SIAM Journal on Numerical Analysis (publishedVersion)
- Repository KITopen (Karlsruhe Institute of Technology) (publishedVersion)
- KITopen (other-oa)
- Topics
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Matrix Theory and Algorithms
- Physical Sciences › Mathematics › Numerical Analysis › Numerical Solvers for Ordinary Differential Equations
- Physical Sciences › Engineering › Computational Mechanics › Advanced Numerical Methods in Computational Mathematics
- Keywords
- matrix exponential, Krylov subspace methods, Arnoldi method, Lanczos method, functional calculus, error bounds, biconjugate gradient method, numerical range, pseudospectra, superlinear convergence, stiff ordinary differential equations, time integration methods, nonlinear differential equations, Jacobian matrix
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.