Quasi-Monte Carlo Methods for some Linear Algebra Problems. Convergence and Complexity

Abstract

We present quasi-Monte Carlo analogs of Monte Carlo methods for some linear algebra problems: solving systems of linear equations, computing extreme eigenvalues, and matrix inversion. Reformulating the problems as solving integral equations with a special kernels and domains permits us to analyze the quasi-Monte Carlo methods with bounds from numerical integration. Standard Monte Carlo methods for integration provide a convergence rate of O(N^(−1/2)) using N samples. Quasi-Monte Carlo methods use quasirandom sequences with the resulting convergence rate for numerical integration as good as O((logN)^k)N^(−1)). We have shown theoretically and through numerical tests that the use of quasirandom sequences improves both the magnitude of the error and the convergence rate of the considered Monte Carlo methods. We also analyze the complexity of considered quasi-Monte Carlo algorithms and compare them to the complexity of the analogous Monte Carlo and deterministic algorithms.

Novel Approach for Eigenvalue Problems Using the Monte Carlo Method

“…Monte Carlo algorithms correspondingly use the concept of the power method combined by the given matrix, the resolvent matrix and the reverse matrix with Monte Carlo iterations [1][2][3][4][5][6][7][8]. Several authors have worked on Advancement of Monte Carlo methods [9][10][11][12][13][14][15][16][17][18][19][20].…”