A fast method to estimate the Moore-Penrose inverse for well-determined numerical rank matrices based on the Tikhonov regularization
Authors
P. Soto-Quiros
- Escuela de Matemática, Instituto Tecnológico de Costa Rica, Cartago 30101, Costa Rica.
Abstract
This paper introduces a novel approach for estimating the Moore-Penrose inverse. The method proposed relies on Tikhonov regularization, which requires the computation of all positive singular values of an \(m \times n\) matrix. Additionally, we present a highly efficient and accurate procedure for estimating these singular values. This procedure assumes the well-determined numerical rank of matrices \(A^*A\) (if \(m \geq n\)) and \(AA^*\) (if \(m \leq n\)). Furthermore, we demonstrate the application of our proposed method in solving linear discrete well-posed problems. The paper concludes with numerical simulations to illustrate the advantages of our novel approach. Notably, we compare the execution time associated with our technique to that of some relevant methods in the existing literature, demonstrating that our method outperforms others in terms of computational efficiency. To further substantiate our findings, we conduct computational experiments to measure execution time and speedup. The results affirm the efficiency of our proposed method, showcasing reduced execution times compared to other methods. This contributes to establishing our approach's practical viability and effectiveness in diverse applications.
Share and Cite
ISRP Style
P. Soto-Quiros, A fast method to estimate the Moore-Penrose inverse for well-determined numerical rank matrices based on the Tikhonov regularization, Journal of Mathematics and Computer Science, 37 (2025), no. 1, 59--81
AMA Style
Soto-Quiros P., A fast method to estimate the Moore-Penrose inverse for well-determined numerical rank matrices based on the Tikhonov regularization. J Math Comput SCI-JM. (2025); 37(1):59--81
Chicago/Turabian Style
Soto-Quiros, P.. "A fast method to estimate the Moore-Penrose inverse for well-determined numerical rank matrices based on the Tikhonov regularization." Journal of Mathematics and Computer Science, 37, no. 1 (2025): 59--81
Keywords
- Moore-Penrose inverse
- Tikhonov regularization
- singular values
- well-determined numerical rank
MSC
- 65F15
- 65F20
- 65F22
- 65F45
- 65H10
References
-
[1]
Z. Allen-Zhu, Y. Li, LazySVD: Even faster SVD decomposition yet without agonizing pain, Adv. Neural Inf. Process Syst., 26 (2019), 9 pages
-
[2]
P. Arbenz, D. Kressner, D. Zürich, Lecture notes on solving large scale eigenvalue problems, D-MATH, EHT Zur., 2 (2012), 256 pages
-
[3]
Armadillo, Armadillo reference documentation of command rank, , (Accessed 25 January 2024),
-
[4]
S. Artidiello, A. Cordero, J. R. Torregrosa, M. P. Vassileva, Generalized inverses estimations by means of iterative methods with memory, Mathematics, 8 (2020), 13 pages
-
[5]
A. Ataei, Improved Qrginv algorithm for computing Moore-Penrose inverse matrices, ISRN Appl. Math., 2014 (2014), 5 pages
-
[6]
J. C. A. Barata, M. S. Hussein, The Moore–Penrose pseudoinverse: A tutorial review of the theory, Braz. J. Phys., 42 (2012), 146–165
-
[7]
A. Ben-Israel, An iterative method for computing the generalized inverse of an arbitrary matrix, Math. Comput., 19 (1965), 452–455
-
[8]
M. W. Benson, P. O. Frederickson, Fast parallel algorithms for the Moore-Penrose pseudo-inverse, SIAM, Philadelphia, PA (1987)
-
[9]
T. Boros, P. Rózsa, An explicit formula for singular values of the Sylvester-Kac matrix, Linear Algebra Appl., 421 (2007), 407–416
-
[10]
M. Brand, Fast low-rank modifications of the thin singular value decomposition, Linear Algebra Appl., 415 (2006), 20–30
-
[11]
D. Calvetti, L. Reichel, Tikhonov regularization of large linear problems, BIT, 43 (2003), 263–283
-
[12]
D. Calvetti, L. Reichel, Tikhonov regularization with a solution constraint, SIAM J. Sci. Comput., 26 (2004), 224–239
-
[13]
J. Chavarría-Molina, J. J. Fallas-Monge, P. Soto-Quiros, Effective implementation to reduce execution time of a low-rank matrix approximation problem, J. Comput. Appl. Math., 401 (2022), 17 pages
-
[14]
H. Chen, Y. Wang, A family of higher-order convergent iterative methods for computing the Moore-Penrose inverse, Appl. Math. Comput., 218 (2011), 4012–4016
-
[15]
J. Chung, M. Chung, Computing optimal low-rank matrix approximations for image processing, In: 2013 Asilomar Conference on Signals, Systems and Computers, IEEE, (2013), 670–674
-
[16]
J. Chung, M. Chung, An efficient approach for computing optimal low-rank regularized inverse matrices, Inverse Probl., 30 (2014), 19 pages
-
[17]
A. Cordero, P. Soto-Quiros, J. R. Torregrosa, A general class of arbitrary order iterative methods for computing generalized inverses, Appl. Math. Comput., 409 (2021), 18 pages
-
[18]
P. Courrieu, Fast computation of Moore-Penrose inverse matrices, Neural Inform. Process. Lett. Rev., 8 (2005), 25–29
-
[19]
H. Cragon, Computer architecture and implementation, Cambridge University Press, (2000)
-
[20]
B. N. Datta, Numerical linear algebra and applications, Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA (2010)
-
[21]
P. Drineas, M. Mahoney, Approximating a Gram matrix for improved kernel-based learning (extended abstract), In: Lecture Notes in Comput. Sci., Springer, Berlin, 3559 (2005), 323–337
-
[22]
M. Elouafi, An eigenvalue localization theorem for pentadiagonal symmetric Toeplitz matrices, Linear Algebra Appl., 435 (2011), 2986–2998
-
[23]
A.-J. Gallego, A. Pertusa, P. Gil, Automatic ship classification from optical aerial images with convolutional neural networks, Remote Sens., 10 (2018), 20 pages
-
[24]
G. H. Golub, M. Heath, G. Wahba, Generalized cross-validation as a method for choosing a good ridge parameter, Technometrics, 21 (1979), 215–223
-
[25]
T. N. E. Greville, Some applications of the pseudoinverse of a matrix, SIAM Rev., 2 (1960), 15–22
-
[26]
P. Guo, D. Zhao, M. Han, S. Feng, Pseudoinverse learners: New trend and applications to big data, In: INNS big data and deep learning conference, Springer, Cham, (2019), 158–168
-
[27]
J. Gutiérrez-Gutiérrez, Eigenvalue decomposition for persymmetric Hankel matrices with at most three non-zero antidiagonals, Appl. Math. Comput., 234 (2014), 333–338
-
[28]
P. C. Hansen, The truncated SVD as a method for regularization, BIT, 27 (1987), 534–553
-
[29]
N. J. Higham, Cholesky factorization, WIREs Comput. Stat., 1 (2009), 251–254
-
[30]
J. T. Holodnak, I. C. F. Ipsen, Randomized approximation of the Gram matrix: exact computation and probabilistic bounds, SIAM J. Matrix Anal. Appl., 36 (2015), 110–137
-
[31]
R. Hunger, Floating point operations in matrix-vector calculus, Tech. Univ. München, (2007)
-
[32]
Julia, Julia reference documentation of command rank, , (Accessed 25 January 2024),
-
[33]
V. N. Katsikis, D. Pappas, Fast computing of the Moore-Penrose inverse matrix, Electron. J. Linear Algebra, 17 (2008), 637–650
-
[34]
V. N. Katsikis, D. Pappas, A. Petralias, An improved method for the computation of the Moore–Penrose inverse matrix, Appl. Math. Comput., 217 (2011), 9828–9834
-
[35]
S. López-Tapia, A. Lucas, R. Molina, A. K. Katsaggelos, A single video super-resolution gan for multiple downsampling operators based on pseudo-inverse image formation models, Digit. Signal Process., 104 (2020), 12 pages
-
[36]
S. Lu, X. Wang, G. Zhang, X. Zhou, Effective algorithms of the Moore-Penrose inverse matrices for extreme learning machine, Intell. Data Anal., 19 (2015), 743–760
-
[37]
T. Lyche, Numerical linear algebra and matrix factorizations, Springer, Cham (2020)
-
[38]
J. Maddock, C. Kormanyos, Boost multiprecision, , (Accessed 25 January 2024),
-
[39]
O. Makkonen, C. Hollanti, Secure distributed gram matrix multiplication, In: 2023 IEEE Information Theory Workshop (ITW), (2023), 192–197
-
[40]
MathWorks, Distributing arrays to parallel workers, , (Accessed 25 January 2024),
-
[41]
MathWorks, Use distributed arrays to solve systems of linear equations with direct methods, , (Accessed 25 January 2024),
-
[42]
MathWorks, Matlab reference documentation of command rank, , (Accessed 25 January 2024),
-
[43]
MathWorks, Matlab reference documentation of command mtimes, , (Accessed 25 January 2024),
-
[44]
MathWorks, Matlab reference documentation of command isillconditioned, , (Accessed 25 January 2024),
-
[45]
S. Miljkovi´c, M. Miladinovi´c, P. Stanimirovi´c, I. Stojanovi´c, Application of the pseudoinverse computation in reconstruction of blurred images, Filomat, 26 (2012), 453–465
-
[46]
NumPy, Numpy reference documentation of command rank, , (Accessed 25 January 2024),
-
[47]
Octave, Octave reference documentation of command rank, , (Accessed 25 January 2024),
-
[48]
M. Ostilli, Cayley trees and Bethe lattices: A concise analysis for mathematicians and physicists, Phys. A, 391 (2012), 3417–3423
-
[49]
R. Penrose, On best approximate solutions of linear matrix equations, Math. Proc. Camb. Philos. Soc., 52 (1956), 17–19
-
[50]
M. D. Petkovi´, Generalized schultz iterative methods for the computation of outer inverses, Comput. Math. Appl., 67 (2014), 1837–1847
-
[51]
M. D. Petkovi´c, M. A. Krsti´c, K. P. Rajkovi´, Rapid generalized schultz iterative methods for the computation of outer inverses, J. Comput. Appl. Math., 344 (2018), 572–584
-
[52]
M. D. Petkovi´c, M. S. Petkovi´c, Hyper-power methods for the computation of outer inverses, J. Comput. Appl. Math., 278 (2015), 110–118
-
[53]
V. Rajaraman, IEEE standard for floating point numbers, Resonance, 21 (2016), 11–30
-
[54]
O. Rojo, M. Robbiano, An explicit formula for eigenvalues of Bethe trees and upper bounds on the largest eigenvalue of any tree, Lin. Algebra Appl., 427 (2007), 138–150
-
[55]
C. Sanderson, R. Curtin, Armadillo: a template-based C++ library for linear algebra, J. Open Source Softw., 1 (2016), 2 pages
-
[56]
Scilab, Scilab reference documentation of command rank, , (Accessed 25 January 2024),
-
[57]
P. Soto-Quiros, J. Jose Fallas-Monge, J. Chavarría-Molina, A fast algorithm for image deconvolution based on a rank constrained inverse matrix approximation problem, In: Proceedings of Sixth International Congress on Information and Communication Technology, Springer, Singapore, (2022), 165–176
-
[58]
P. S. Stanimirovi´c, D. Pappas, V. N. Katsikis, I. P. Stanimirovi´c, Full-rank representations of outer inverses based on the QR decomposition, Appl. Math. Comput., 218 (2012), 10321–10333
-
[59]
A. T˘an˘asescu, P. G. Popescu, A fast singular value decomposition algorithm of general k-tridiagonal matrices, J. Comput. Sci., 31 (2019), 1–5
-
[60]
B. Telfer, D. Casasent, Fast method for updating robust pseudoinverse and Ho-Kashyap associative processors, IEEE Trans. Syst. Man. Cybern., 24 (1994), 1387–1390
-
[61]
A. Torokhti, P. Soto-Quiros, Generalized Brillinger-like transforms, IEEE Signal Process. Lett., 23 (2016), 843–847
-
[62]
G. Wang, Y. Wei, S. Qiao, Generalized inverses: Theory and computations, Springer, Singapore; Science Press Beijing, Beijing (2018)
-
[63]
J. Wang, P. Guo, X. Xin, Review of pseudoinverse learning algorithm for multilayer neural networks and applications, In: Advances in Neural Networks – ISNN 2018, Lecture Notes in Computer Science, Springer, Cham, (2018), 99–106
-
[64]
R. E. White, Computational Linear Algebra: with Applications and MATLAB® Computations, Chapman & Hall/CRC, New York (2023)
-
[65]
L. Wu, Regularization methods and algorithms for least squares and Kronecker product least squares problems, ProQuest LLC, Ann Arbor, MI (1997)
-
[66]
W. Xu, S. Qiao, A fast symmetric SVD algorithm for square Hankel matrices, Linear Algebra Appl., 428 (2008), 550–563
-
[67]
C. Xu, K. Wang, Z. Liu, Singular value decomposition for central extended matrix, In: 2011 International Conference on Computer Science and Service System (CSSS), IEEE, (2011), 736–738
-
[68]
G. Zielke, Report on test matrices for generalized inverses, Computing, 36 (1986), 105–162
-
[69]
D. D. Zontini, M. L. Mirkoski, J. A. Santos, Error bounds in the computation of outer inverses with generalized Schultz iterative methods and its use in computing of Moore-Penrose inverse, Appl. Math. Comput., 440 (2023), 10 pages
