Stochastic Gradient Descent for Linear Systems with Missing Data

Year:    2019

Numerical Mathematics: Theory, Methods and Applications, Vol. 12 (2019), Iss. 1 : pp. 1–20

Abstract

Traditional methods for solving linear systems have quickly become impractical due to an increase in the size of available data. Utilizing massive amounts of data is further complicated when the data is incomplete or has missing entries. In this work, we address the obstacles presented when working with large data and incomplete data simultaneously. In particular, we propose to adapt the Stochastic Gradient Descent method to address missing data in linear systems. Our proposed algorithm, the Stochastic Gradient Descent for Missing Data method (mSGD), is introduced and theoretical convergence guarantees are provided. In addition, we include numerical experiments on simulated and real world data that demonstrate the usefulness of our method.

Journal Article Details

Publisher Name:    Global Science Press

Language:    English

DOI:    https://doi.org/10.4208/nmtma.OA-2018-0066

Numerical Mathematics: Theory, Methods and Applications, Vol. 12 (2019), Iss. 1 : pp. 1–20

Published online:    2019-01

AMS Subject Headings:   

Copyright:    COPYRIGHT: © Global Science Press

Pages:    20

Keywords:    Linear systems missing data iterative methods least squares problems.

  1. Randomized Douglas–Rachford Methods for Linear Systems: Improved Accuracy and Efficiency

    Han, Deren | Su, Yansheng | Xie, Jiaxin

    SIAM Journal on Optimization, Vol. 34 (2024), Iss. 1 P.1045

    https://doi.org/10.1137/23M1567503 [Citations: 5]
  2. Adaptive Deep Incremental Learning — Assisted Missing Data Imputation for Streaming Data

    Syavasya, C. V. S. R. | Lakshmi, M. A.

    Journal of Interconnection Networks, Vol. 22 (2022), Iss. Supp02

    https://doi.org/10.1142/S021926592143009X [Citations: 1]
  3. Quantile-based Random Kaczmarz for corrupted linear systems of equations

    Steinerberger, Stefan

    Information and Inference: A Journal of the IMA, Vol. 12 (2023), Iss. 1 P.448

    https://doi.org/10.1093/imaiai/iaab029 [Citations: 7]
  4. Approximate Solutions of Linear Systems at a Universal Rate

    Steinerberger, Stefan

    SIAM Journal on Matrix Analysis and Applications, Vol. 44 (2023), Iss. 3 P.1436

    https://doi.org/10.1137/22M1517196 [Citations: 0]
  5. Randomized Kaczmarz method with adaptive stepsizes for inconsistent linear systems

    Zeng, Yun | Han, Deren | Su, Yansheng | Xie, Jiaxin

    Numerical Algorithms, Vol. 94 (2023), Iss. 3 P.1403

    https://doi.org/10.1007/s11075-023-01540-x [Citations: 4]
  6. Surrounding the solution of a linear system of equations from all sides

    Steinerberger, Stefan

    Quarterly of Applied Mathematics, Vol. 79 (2021), Iss. 3 P.419

    https://doi.org/10.1090/qam/1587 [Citations: 8]
  7. On Adaptive Stochastic Heavy Ball Momentum for Solving Linear Systems

    Zeng, Yun | Han, Deren | Su, Yansheng | Xie, Jiaxin

    SIAM Journal on Matrix Analysis and Applications, Vol. 45 (2024), Iss. 3 P.1259

    https://doi.org/10.1137/23M1575883 [Citations: 1]