Supplementary directional relaxations for the acceleration of Kaczmarz's projection method
DOI:
https://doi.org/10.33993/jnaat321-739Keywords:
least-squares problems, Kaczmarz's projection algorithm, supplementary projectionsAbstract
Starting from an extension of Kaczmarz's method, obtained by us in a previous paper, we introduce new directions for projections. We prove that by this, we don't modify the set of limit points of the original extended Kaczmarz algorithm. For the class of boundary value problems or integral equations of the first kind, we describe a method for constructing these new directions. It is based on considering coarser level of discretization for the initial problem. Some numerical experiments are also presented.Downloads
References
Censor, Y., Row-action methods for huge and sparse systems and their applications, SIAM Review, 23, no. 4, pp. 444-466, 1981, https://doi.org/10.1137/1023097. DOI: https://doi.org/10.1137/1023097
Censor, Y. and Stavros A. Z., Parallel optimization: theory, algorithms and applications, "Numer. math. and Sci. Comp." Series, Oxford Univ. Press, New York, 1997.
Engl., H. W., Regularization methods for the stable solution of inverse problems, Surv. Math. Ind., 3, pp. 71-143, 1993.
Golub, G. H. and van Loan, C. F., Matrix computations, The John's Hopkins Univ. Press, Baltimore, 1983.
Griebel, M., Multilevel algorithms considered as iterative methods on semidefinite systems, SIAM J. Sci. Comp., 15, no. 3, pp. 547-565, 1994, https://doi.org/10.1137/0915036. DOI: https://doi.org/10.1137/0915036
Kaczmarz, S., Angenaherte Auflosung von Systemen linearer Gleischungen, Bull. Acad. Polonaise Sci. et Lettres A, pp. 355-357, 1937.
Kress, R., Linear integral equations, Springer-Verlag, Berlin, 1989. DOI: https://doi.org/10.1007/978-3-642-97146-4
Popa, C., Least-squares solutions of overdetermined inconsistent linear systems using Kaczmarz's relaxation, Inter. J. Comp. Math., 55, pp. 79-89, 1995, https://doi.org/10.1080/00207169508804364. DOI: https://doi.org/10.1080/00207169508804364
Popa, C., Extensions of block-projections methods with relaxation parameters to inconsistent and rank-defficient least-squares problems, BIT, 38, no. 1, pp. 151-176, 1998, https://doi.org/10.1007/BF02510922. DOI: https://doi.org/10.1007/BF02510922
Popa, C., Characterization of the solutions set of inconsistent least-squares problems by an extended Kaczmarz algorithm, Korean J. Comp. Appl. Math., 6, no. 1, pp. 51-64, 1999, https://doi.org/10.1007/BF02941906. DOI: https://doi.org/10.1007/BF02941906
Popa, C., Preconditioned Kaczmarz-extended algorithm with relaxation parameters, Korean J. Comp. Appl. Math., 8, no. 1, pp. 9-26, 2000, https://doi.org/10.1007/BF03009946. DOI: https://doi.org/10.1007/BF03009946
Popa, C., Direct and iterative Kaczmarz-like solvers, submitted for publication, 2002.
Tanabe, K., Projection method for solving a singular system of linear equations and its applications, Numer. Math., 17, pp. 203-214, 1971, https://doi.org/10.1007/BF01436376. DOI: https://doi.org/10.1007/BF01436376
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2015 Journal of Numerical Analysis and Approximation Theory
This work is licensed under a Creative Commons Attribution 4.0 International License.
Open Access. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.