Abstract
The randomized Kaczmarz algorithm is a simple iterative method for solving linear systems of equations. This study proposes a variant of the randomized Kaczmarz algorithm by combining block projection and weighted averaging techniques. Here, block projection quickly decreases iterative errors, and averaging reduces randomness and enables parallel computation simultaneously. Their combination can balance the convergence rate, convergence horizon, and computational complexity. In addition, three adaptive weights are designed to balance multiple block calculations and accelerate the proposed method. Exponential convergence is established for general linear systems (overdetermined or underdetermined, full-rank or deficient-rank, and consistent or inconsistent). Numerical simulations explain and verify the results.
Similar content being viewed by others
References
Bai, Z.Z., Wu, W.T.: On convergence rate of the randomized Kaczmarz method. Linear Algebra Appl. 553, 252–269 (2018). https://doi.org/10.1016/j.laa.2018.05.009
Bai, Z.Z., Wu, W.T.: On greedy randomized Kaczmarz method for solving large sparse linear systems. SIAM J. Sci. Comput. 40(1), A592–A606 (2018). https://doi.org/10.1137/17m1137747
Bai, Z.Z., Wu, W.T.: On greedy randomized augmented Kaczmarz method for solving large sparse inconsistent linear systems. SIAM J. Sci. Comput. 43(6), A3892–A3911 (2021). https://doi.org/10.1137/20m1352235
Censor, Y., Eggermont, P.P.B., Gordon, D.: Strong underrelaxation in Kaczmarz’s method for inconsistent systems. Numer. Math. 41(1), 83–92 (1983). https://doi.org/10.1007/bf01396307
Chen, J.Q., Huang, Z.D.: On a fast deterministic block Kaczmarz method for solving large-scale linear systems. Numer. Algorithms (2021). https://doi.org/10.1007/s11075-021-01143-4
Du, K., Si, W.T., Sun, X.H.: Randomized extended average block Kaczmarz for solving least squares. SIAM J. Sci. Comput. 42(6), A3541–A3559 (2020). https://doi.org/10.1137/20m1312629
Dumitrescu, B.: On the relation between the randomized extended Kaczmarz algorithm and coordinate descent. BIT Numer. Math. 55(4), 1005–1015 (2014). https://doi.org/10.1007/s10543-014-0526-9
Eggermont, P., Herman, G., Lent, A.: Iterative algorithms for large partitioned linear systems, with applications to image reconstruction. Linear Algebra Appl. 40, 37–67 (1981). https://doi.org/10.1016/0024-3795(81)90139-7
Gordon, D., Gordon, R.: Component-averaged row projections: a robust, block-parallel scheme for sparse linear systems. SIAM J. Sci. Comput. 27(3), 1092–1117 (2005). https://doi.org/10.1137/040609458
Gordon, D., Gordon, R.: Carp-cg: a robust and efficient parallel solver for linear systems, applied to strongly convection dominated PDEs. Parallel Comput. 36(9), 495–515 (2010). https://doi.org/10.1016/j.parco.2010.05.004
Gower, R.M., Richtárik, P.: Randomized iterative methods for linear systems. SIAM J. Matrix Anal. Appl. 36(4), 1660–1690 (2015). https://doi.org/10.1137/15m1025487
Gower, R.M., Molitor, D., Moorman, J., Needell, D.: On adaptive sketch-and-project for solving linear systems. SIAM J. Matrix Anal. Appl. 42(2), 954–989 (2021). https://doi.org/10.1137/19m1285846
Haddock, J., Needell, D.: On Motzkin’s method for inconsistent linear systems. BIT Numer. Math. 59(2), 387–401 (2018). https://doi.org/10.1007/s10543-018-0737-6
Kaczmarz, S.: Approximate solution of systems of linear equations. Int. J. Control 57(6), 1269–1271 (1993). https://doi.org/10.1080/00207179308934446
Kamath, G., Ramanan, P., Song, W.Z.: Distributed randomized Kaczmarz and applications to seismic imaging in sensor network. In: 2015 International Conference on Distributed Computing in Sensor Systems, pp. 169–178 (2015). https://doi.org/10.1109/DCOSS.2015.27
Li, J., Hu, Z.C.: Toeplitz lemma, complete convergence, and complete moment convergence. Commun. Stat. Theory Methods 46(4), 1731–1743 (2017)
Lin, J., Zhou, D.X.: Learning theory of randomized Kaczmarz algorithm. J. Mach. Learn. Res. 16(103), 3341–3365 (2015)
Liu, Y., Gu, C.Q.: On greedy randomized block Kaczmarz method for consistent linear systems. Linear Algebra Appl. 616, 178–200 (2021). https://doi.org/10.1016/j.laa.2021.01.024
Loera, J.A.D., Haddock, J., Needell, D.: A sampling Kaczmarz–Motzkin algorithm for linear feasibility. SIAM J. Sci. Comput. 39(5), S66–S87 (2017). https://doi.org/10.1137/16m1073807
Lorenz, D.A., Rose, S., Schöpfer, F.: The randomized Kaczmarz method with mismatched adjoint. BIT Numer. Math. 58(4), 1079–1098 (2018). https://doi.org/10.1007/s10543-018-0717-x
Ma, A., Molitor, D.: Randomized Kaczmarz for tensor linear systems. BIT Numer. Math. 62(1), 171–194 (2021). https://doi.org/10.1007/s10543-021-00877-w
Moorman, J.D., Tu, T.K., Molitor, D., Needell, D.: Randomized Kaczmarz with averaging. BIT Numer. Math. 61(1), 337–359 (2020). https://doi.org/10.1007/s10543-020-00824-1
Morijiri, Y., Aishima, K., Matsuo, T.: Extension of an error analysis of the randomized Kaczmarz method for inconsistent linear systems. JSIAM Lett. 10, 17–20 (2018). https://doi.org/10.14495/jsiaml.10.17
Necoara, I.: Faster randomized block Kaczmarz algorithms. SIAM J. Matrix Anal. Appl. 40(4), 1425–1452 (2019). https://doi.org/10.1137/19m1251643
Needell, D.: Randomized Kaczmarz solver for noisy linear systems. BIT Numer. Math. 50(2), 395–403 (2010). https://doi.org/10.1007/s10543-010-0265-5
Needell, D., Tropp, J.A.: Paved with good intentions: Analysis of a randomized block Kaczmarz method. Linear Algebra Appl. 441, 199–221 (2014). https://doi.org/10.1016/j.laa.2012.12.022
Needell, D., Ward, R.: Batched stochastic gradient descent with weighted sampling. arXiv:1608.07641 (2017)
Needell, D., Srebro, N., Ward, R.: Stochastic gradient descent, weighted sampling, and the randomized Kaczmarz algorithm. Math. Program. 155(1–2), 549–573 (2015). https://doi.org/10.1007/s10107-015-0864-7
Needell, D., Zhao, R., Zouzias, A.: Randomized block Kaczmarz method with projection for solving least squares. Linear Algebra Appl. 484, 322–343 (2015). https://doi.org/10.1016/j.laa.2015.06.027
Pantelimon, I., Popa, C.: Constraining by a family of strictly nonexpansive idempotent functions with applications in image reconstruction. BIT Numer. Math. 53, 527–544 (2013). https://doi.org/10.1007/s10543-012-0414-0
Rebrova, E., Needell, D.: On block Gaussian sketching for the Kaczmarz method. Numer. Algorithms 86(1), 443–473 (2020). https://doi.org/10.1007/s11075-020-00895-9
Richtárik, P., Takáč, M.: Stochastic reformulations of linear systems: algorithms and convergence theory. SIAM J. Matrix Anal. Appl. 41(2), 487–524 (2020). https://doi.org/10.1137/18m1179249
Steinerberger, S.: Randomized Kaczmarz converges along small singular vectors. SIAM J. Matrix Anal. Appl. 42(2), 608–615 (2021). https://doi.org/10.1137/20m1350947
Steinerberger, S.: A weighted randomized Kaczmarz method for solving linear systems. Math. Comput. 90(332), 2815–2826 (2021). https://doi.org/10.1090/mcom/3644
Strohmer, T., Vershynin, R.: A randomized Kaczmarz algorithm with exponential convergence. J. Fourier Anal. Appl. 15(2), 262–278 (2008). https://doi.org/10.1007/s00041-008-9030-4
Yi, P., Lei, J., Chen, J., Hong, Y., Shi, G.: Distributed linear equations over random networks. IEEE Trans. Autom. Control (2022). https://doi.org/10.1109/TAC.2022.3187379
Zeng, Y., Han, D., Su, Y., Xie, J.: Randomized Kaczmarz method with adaptive stepsizes for inconsistent linear systems. Numer. Algorithms (2023). https://doi.org/10.1007/s11075-023-01540-x
Zouzias, A., Freris, N.M.: Randomized extended Kaczmarz for solving least squares. SIAM J. Matrix Anal. Appl. 34(2), 773–793 (2013). https://doi.org/10.1137/120889897
Zouzias, A., Freris, N.M.: Randomized gossip algorithms for solving Laplacian systems. In: 2015 European Control Conference (ECC), pp. 1920–1925 (2015). https://doi.org/10.1109/ECC.2015.7330819
Acknowledgements
We would like to thank the anonymous reviewers for their valuable comments, which have simplified the proofs and tremendously improved the content and quality of this paper.
Author information
Authors and Affiliations
Contributions
Both authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Zeyi Zhang and Dong Shen. The first draft of the manuscript was written by Zeyi Zhang and both authors prepared the revised version. Both authors approved the final submission.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no conflict of interest.
Additional information
Communicated by Gunnar Martinsson.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was supported by the Beijing Natural Science Foundation (Z210002) and National Natural Science Foundation of China [62173333].
Appendices
Appendices
Proof of Lemma 3.1
For a real number sequence \(\{a_i\}\), we can verify that
As k goes to infinity, the left term in (7.1) converges to one.
Denote \(\varPhi (k+1,t)= (1-a_k) \varPhi (k,t)\), \(\varPhi _{t,t}=1\), \(\forall k\ge t\). Using (7.1), we have
It is easy to see that \(\lim _{k\rightarrow \infty }\varPhi (k,t)=0\), \(\forall t\). Applying the Toeplitz Lemma [16] to the second term on the right-hand side of (7.2), we obtain that \(x_{k}\) converges to \(\lim _{k\rightarrow \infty } \frac{b_k}{a_k}\) if the limit exists. \(\square \)
Proof of Theorem 3.5
From (3.5), we have
The disturbed term is
\(\square \)
Note that
where one should notice that \({\textbf{A}}^j_{\{i\}}({\textbf{A}}^j_{\{i\}})^T\) is a nonzero number because \({\textbf{A}}^j_{\{i\}}\) is a nonzero row vector.
Taking expectation on the disturbed term, we have
Finally, note that \({\textbf{X}}_j=({\textbf{A}}^j)^T{\textbf{Z}}_j{\textbf{A}}^j\), \(\omega _j=\Vert {\textbf{A}}^j\Vert _F ^2/\Vert {\textbf{A}}\Vert _F ^2\), we have
Then, if \(\alpha \le 1\), we obtain \(\Vert {\mathbb {E}}[{{\textbf {x}}}_k-{{\textbf {x}}}_*]\Vert \le (1-\frac{\alpha \sigma _r({{\textbf {A}}}^T{{\textbf {A}}})}{\Vert {{\textbf {A}}}\Vert _F^2})\Vert {\mathbb {E}}[{{\textbf {x}}}_{k-1}-{{\textbf {x}}}_*]\Vert \).
Proof of Lemma 4.1
Denote
Then, we obtain
\(\square \)
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhang, Z., Shen, D. Randomized Kaczmarz algorithm with averaging and block projection. Bit Numer Math 64, 1 (2024). https://doi.org/10.1007/s10543-023-01002-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10543-023-01002-9