Abstract
QR factorization is a key tool in mathematics, computer science, operations research, and engineering. This paper presents the roundoff-error-free (REF) QR factorization framework comprising integer-preserving versions of the standard and the thin QR factorizations and associated algorithms to compute them. Specifically, the standard REF QR factorization factors a given matrix \(A\in {\mathbb {Z}}^{m\times n}\) as \(A=QDR\), where \(Q\in {\mathbb {Z}}^{m\times m}\) has pairwise orthogonal columns, D is a diagonal matrix, and \(R\in {\mathbb {Z}}^{m\times n}\) is an upper trapezoidal matrix; notably, the entries of Q and R are integral, while the entries of D are reciprocals of integers. In the thin REF QR factorization, \(Q\in {\mathbb {Z}}^{m\times n}\) also has pairwise orthogonal columns, and \(R\in {\mathbb {Z}}^{n\times n}\) is also an upper triangular matrix. In contrast to traditional (i.e., floating-point) QR factorizations, every operation used to compute these factors is integral; thus, REF QR is guaranteed to be an exact orthogonal decomposition. Importantly, the bit-length of every entry in the REF QR factorizations (and within the algorithms to compute them) is bounded polynomially. Notable applications of our REF QR factorizations include finding exact least squares or exact basic solutions, \({\textbf{x}}\in {\mathbb {Q}}^n\), to any given full column rank or rank deficient linear system \(A {\textbf{x}}= {\textbf{b}}\), respectively. In addition, our exact factorizations can be used as a subroutine within exact and/or high-precision quadratic programming. Altogether, REF QR provides a framework to obtain exact orthogonal factorizations of any rational matrix (as any rational/decimal matrix can be easily transformed into an integral matrix).
Similar content being viewed by others
Data availability
Data sharing does not apply to this article as no datasets were generated or analyzed during this study.
Notes
Throughout the paper, the symbol \(\triangleq \) means “is defined as.”
References
Bauer, F.L.: Elimination with weighted row combinations for solving linear equations and least squares problems. Numer. Math. 7(4), 338–352 (1965). https://doi.org/10.1007/BF01436528
Golub, G.H., Wilkinson, J.H.: Note on the iterative refinement of least squares solution. Numer. Math. 9(2), 139–148 (1966)
Chan, T.F.: Rank revealing QR factorizations. Linear Algebra Appl. 88(C), 67–82 (1987). https://doi.org/10.1016/0024-3795(87)90103-0
Chan, T.F., Hansen, P.C.: Some applications of the rank revealing QR factorization. Stat. Comput. 13(3), 727–741 (1992)
Higham, N.J.: QR factorization with complete pivoting and accurate computation of the SVD. Linear Algebra Appl. 309(1–3), 153–174 (2000)
Higham, N.J.: Accuracy and Stability of Numerical Algorithms, 2nd edn. Siam, Philadelphia, PA, USA (2002)
Fukaya, T., Kannan, R., Nakatsukasa, Y., Yamamoto, Y., Yanagisawa, Y.: Shifted Cholesky QR for computing the QR factorization of ill-conditioned matrices. SIAM J. Sci. Comput. 42(1), 477–503 (2020). https://doi.org/10.1137/18M1218212. arXiv:1809.11085
Francis, J.G.F.: The QR transformation a unitary analogue to the LR transformation-part 1. Comput. J. 4(3), 265–271 (1961). https://doi.org/10.1093/comjnl/4.3.265
Kublanovskaya, V.N.: On some algorithms for the solution of the complete eigenvalue problem. USSR Comput. Math. Math. Phys. 1(3), 637–657 (1962). https://doi.org/10.1016/0041-5553(63)90168-X
Parlett, B.N., Poole, W.G., Jr.: A geometric theory for the QR, LU and power iterations. SIAM J. Numer. Anal. 10(2), 389–412 (1973)
Gill, P.E., Murray, W.: Numerically stable methods for quadratic programming. Math. Program. 14(1), 349–372 (1978). https://doi.org/10.1007/BF01588976
Arioli, M.: The use of QR factorization in sparse quadratic programming and backward error issues. SIAM J. Matrix Anal. Appl. 21(3), 825–839 (2000). https://doi.org/10.1137/S0895479898338147
Weber, T., Sager, S., Gleixner, A.: Solving quadratic programs to high precision using scaled iterative refinement. Math. Program. Comput. 11(3), 421–455 (2019)
Gärtner, B., Schönherr, S.: An efficient, exact, and generic quadratic programming solver for geometric optimization. In: Proceedings of the Sixteenth Annual Symposium on Computational Geometry, pp. 110–118 (2000)
Golub, G.H., Van Loan, C.F.: Matrix Computations. Johns Hopkins Studies in the Mathematical Sciences, vol. 3, 4th edn. Johns Hopkins University Press, Baltimore (2013)
Drmač, Z., Bujanović, Z.: On the failure of rank-revealing QR factorization software: a case study. ACM Trans. Math. Softw. (TOMS) 35(2), 1–28 (2008). https://doi.org/10.1145/1377612.1377616
Zhang, S., Baharlouei, E., Wu, P.: High accuracy matrix computations on neural engines: A study of QR factorization and its applications. In: Proceedings of the 29th International Symposium on High-Performance Parallel and Distributed Computing, pp. 17–28. Association for Computing Machinery, Inc, New York, USA (2020). https://doi.org/10.1145/3369583.3392685
Escobedo, A.R., Moreno-Centeno, E.: Roundoff-error-free algorithms for solving linear systems via Cholesky and LU factorizations. INFORMS J. Comput. 27(4), 677–689 (2015). https://doi.org/10.1287/ijoc.2015.0653
Lourenco, C.J., Moreno-Centeno, E.: Exactly solving sparse rational linear systems via roundoff-error-free Cholesky Factorizations. SIAM J. Matrix Anal. Appl. 43(1), 439–463 (2022). https://doi.org/10.1137/20M1371592
Zhou, W., Jeffrey, D.J.: Fraction-free matrix factors: new forms for LU and QR factors. Front. Comput. Sci. China 2(1), 67–80 (2008). https://doi.org/10.1007/s11704-008-0005-z
Davis, T.A.: Direct Methods for Sparse Linear Systems. SIAM, Philadelphia, PA, USA (2006)
Rice, J.R.: Experiments on Gram–Schmidt orthogonalization. Math. Comput. 20(94), 325–328 (1966)
Givens, W.: Computation of plain unitary rotations transforming a general matrix to triangular form. J. Soc. Ind. Appl. Math. 6(1), 26–50 (1958)
Householder, A.S.: Unitary triangularization of a nonsymmetric matrix. J. ACM (JACM) 5(4), 339–342 (1958)
Pursell, L., Trimble, S.Y.: Gram–Schmidt orthogonalization by Gauss elimination. Am. Math. Mon. 98(6), 544–549 (1991). https://doi.org/10.1080/00029890.1991.11995755
Edmonds, J.: Systems of distinct representatives and linear algebra. J. Res. Nat. Bur. Stand. Sect. B 71(4), 241–245 (1967)
Bareiss, E.H.: Sylvester’s identity and multistep integer-preserving Gaussian elimination. Math. Comput. 22(103), 565–578 (1968)
Montante-Pardo, R.M., Méndez-Cavazos, M.A.: Un método númerico para cálculo matricial. Rev. Técnico-Científica de Divulgación 2, 1–24 (1977)
Lourenco, C., Escobedo, A.R., Moreno-Centeno, E., Davis, T.A.: Exact solution of sparse linear systems via left-looking roundoff-error-free LU factorization in time proportional to arithmetic work. SIAM J. Matrix Anal. Appl. 40(2), 609–638 (2019). https://doi.org/10.1137/18M1202499
Hadamard, J.: Résolution d’une question relative aux déterminants. Bull. Sci. Math. 17(1), 240–246 (1893)
Horn, R.A., Johnson, C.R.: Matrix Analysis. Cambridge University Press, New York (2012)
Davis, T.A., Rajamanickam, S., Sid-Lakhdar, W.M.: A survey of direct methods for sparse linear systems. Acta Numer. 25, 383–566 (2016). https://doi.org/10.1017/S0962492916000076
Foster, L., Kommu, R.: Algorithm 853: an efficient algorithm for solving rank-deficient least squares problems. ACM Trans. Math. Softw. (TOMS) 32(1), 157–165 (2006)
Foster, L.V., Davis, T.: Reliable calculation of numerical rank, null space bases, basic solutions and pseudoinverse solutions using SuiteSparseQR. In: Householder Symposium XVIII on Numerical Linear Algebra, p. 79 (2011)
Golub, G., Pereyra, V.: Differentiation of pseudoinverses, separable nonlinear least squares and other tales. In: Generalized Inverses and Applications, pp. 303–324. Academic Press, New York (1976)
Funding
The first author was partially supported by the USNA JR NARC. The second author was partially supported by NSF under Grant No OAC-1835499.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Lourenco, C., Moreno-Centeno, E. Exact QR factorizations of rectangular matrices. Optim Lett 18, 681–695 (2024). https://doi.org/10.1007/s11590-024-02095-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11590-024-02095-z