A Monte Carlo method for computing the action of a matrix exponential on a vector

A Monte Carlo method for computing the action of a matrix exponential for a certain class of matrices on a vector is proposed. The method is based on generating random paths, which evolve through the indices of the matrix, governed by a given continuous-time Markov chain. The vector solution is computed probabilistically by averaging over a suitable multiplicative functional. This representation extends the existing linear algebra Monte Carlo-based methods, and was used in practice to develop an efficient algorithm capable of computing both, a single entry or the full vector solution. Finally, several relevant benchmarks were executed to assess the performance of the algorithm. A comparison with the results obtained with a Krylov-based method shows the remarkable performance of the algorithm for solving large-scale problems.

[1]  R. Mattheij,et al.  Partial Differential Equations: Modeling, Analysis, Computation (Siam Monographs on Mathematical Modeling and Computation) (Saim Models on Mathematical Modeling and Computation) , 2005 .

[2]  R. A. Leibler,et al.  Matrix inversion by a Monte Carlo method , 1950 .

[3]  V. Alexandrov,et al.  Parallel resolvent Monte Carlo algorithms for linear algebra problems , 2001 .

[4]  M. Sadiku.  Monte Carlo Methods for Electromagnetics , 2009 .

[5]  Sean McKee,et al.  Monte Carlo Methods for Applied Scientists , 2005 .

[6]  J. M. Sellier,et al.  A new Walk on Equations Monte Carlo method for solving systems of linear algebraic equations , 2015 .

[7]  Piero Lanucara,et al.  Domain Decomposition Solution of Elliptic Boundary-Value Problems via Monte Carlo and Quasi-Monte Carlo Methods , 2005, SIAM J. Sci. Comput..

[8]  Giray Ökten,et al.  Solving Linear Equations by Monte Carlo Simulation , 2005, SIAM J. Sci. Comput..

[9]  Marco Caliari,et al.  The Leja Method Revisited: Backward Error Analysis for the Matrix Exponential , 2015, SIAM J. Sci. Comput..

[10]  Paolo Boldi.  TotalRank: ranking without damping , 2005, WWW '05.

[11]  Michele Benzi,et al.  The Physics of Communicability in Complex Networks , 2011, ArXiv.

[12]  Michele Benzi,et al.  MATRIX FUNCTIONS , 2006 .

[13]  Nicholas J. Higham,et al.  Functions of matrices - theory and computation , 2008 .

[14]  Juan A. Acebrón,et al.  Highly efficient numerical algorithm based on random trees for accelerating parallel Vlasov-Poisson simulations , 2013, J. Comput. Phys..

[15]  Michele Benzi,et al.  Analysis of Monte Carlo accelerated iterative methods for sparse linear systems , 2017, Numer. Linear Algebra Appl..

[16]  Nisheeth K. Vishnoi,et al.  Approximating the exponential, the lanczos method and an Õ(m)-time spectral algorithm for balanced separator , 2011, STOC '12.

[17]  W. Stewart,et al.  A numerical study of large sparse matrix exponentials arising in Markov chains 1 1 This work has ben , 1999 .

[18]  M. Pusa,et al.  Computing the Matrix Exponential in Burnup Calculations , 2010 .

[19]  M. Benzi,et al.  Quadrature rule-based bounds for functions of adjacency matrices , 2010 .

[20]  Tim B. Swartz,et al.  Approximating Integrals Via Monte Carlo and Deterministic Methods , 2000 .

[21]  Ivan Todor Dimov,et al.  A new highly convergent Monte Carlo method for matrix computations 1 Supported by The Royal Society , 1998 .

[22]  Yaohang Li,et al.  Convergence Analysis of Markov Chain Monte Carlo Linear Solvers Using Ulam-von Neumann Algorithm , 2013, SIAM J. Numer. Anal..

[23]  Chung-Kuan Cheng,et al.  Power grid simulation using matrix exponential method with rational Krylov subspaces , 2013, 2013 IEEE 10th International Conference on ASIC.

[24]  R. Merris.  Laplacian matrices of graphs: a survey , 1994 .

[25]  M. Eiermann,et al.  Implementation of a restarted Krylov subspace method for the evaluation of matrix functions , 2008 .

[26]  Michele Benzi,et al.  Total communicability as a centrality measure , 2013, J. Complex Networks.

[27]  Awad H. Al-Mohy,et al.  Computing matrix functions , 2010, Acta Numerica.

[28]  C. Lubich,et al.  Error Bounds for Exponential Operator Splittings , 2000 .

[29]  Awad H. Al-Mohy,et al.  Computing the Action of the Matrix Exponential, with an Application to Exponential Integrators , 2011, SIAM J. Sci. Comput..

[30]  Francisco Bernal,et al.  An Efficient Algorithm for Accelerating Monte Carlo Approximations of the Solution to Boundary Value Problems , 2015, Journal of Scientific Computing.