IDEAS home Printed from https://ideas.repec.org/a/spr/joptap/v197y2023i2d10.1007_s10957-023-02193-5.html
   My bibliography  Save this article

Accelerated Doubly Stochastic Gradient Descent for Tensor CP Decomposition

Author

Listed:
  • Qingsong Wang

    (Beihang University)

  • Chunfeng Cui

    (Beihang University)

  • Deren Han

    (Beihang University)

Abstract

In this paper, we focus on the acceleration of doubly stochastic gradient descent method for computing the CANDECOMP/PARAFAC (CP) decomposition of tensors. This optimization problem has N blocks, where N is the order of the tensor. Under the doubly stochastic framework, each block subproblem is solved by the vanilla stochastic gradient method. However, the convergence analysis requires that the variance converges to zero, which is hard to check in practice and may not hold in some implementations. In this paper, we propose accelerating the stochastic gradient method by the momentum acceleration and the variance reduction technique, denoted as DS-MVR. Theoretically, the convergence of DS-MVR only requires the variance to be bounded. Under mild conditions, we show DS-MVR converges to a stochastic $$\varepsilon $$ ε -stationary solution in $$\tilde{\mathcal {O}}(N^{3/2}\varepsilon ^{-3})$$ O ~ ( N 3 / 2 ε - 3 ) iterations with varying stepsizes and in $$\mathcal {O}(N^{3/2}\varepsilon ^{-3})$$ O ( N 3 / 2 ε - 3 ) iterations with constant stepsizes, respectively. Numerical experiments on four real-world datasets show that our proposed algorithm can get better results compared with the baselines.

Suggested Citation

  • Qingsong Wang & Chunfeng Cui & Deren Han, 2023. "Accelerated Doubly Stochastic Gradient Descent for Tensor CP Decomposition," Journal of Optimization Theory and Applications, Springer, vol. 197(2), pages 665-704, May.
  • Handle: RePEc:spr:joptap:v:197:y:2023:i:2:d:10.1007_s10957-023-02193-5
    DOI: 10.1007/s10957-023-02193-5
    as

    Download full text from publisher

    File URL: http://link.springer.com/10.1007/s10957-023-02193-5
    File Function: Abstract
    Download Restriction: Access to the full text of the articles in this series is restricted.

    File URL: https://libkey.io/10.1007/s10957-023-02193-5?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to search for a different version of it.

    References listed on IDEAS

    as
    1. J. Carroll & Jih-Jie Chang, 1970. "Analysis of individual differences in multidimensional scaling via an n-way generalization of “Eckart-Young” decomposition," Psychometrika, Springer;The Psychometric Society, vol. 35(3), pages 283-319, September.
    2. Yangyang Xu & Yibo Xu, 2023. "Momentum-Based Variance-Reduced Proximal Stochastic Gradient Method for Composite Nonconvex Stochastic Optimization," Journal of Optimization Theory and Applications, Springer, vol. 196(1), pages 266-297, January.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Mariela González-Narváez & María José Fernández-Gómez & Susana Mendes & José-Luis Molina & Omar Ruiz-Barzola & Purificación Galindo-Villardón, 2021. "Study of Temporal Variations in Species–Environment Association through an Innovative Multivariate Method: MixSTATICO," Sustainability, MDPI, vol. 13(11), pages 1-25, May.
    2. Wedel, M. & Bijmolt, T.H.A., 1998. "Mixed Tree and Spatial Representation of Dissimilarity Judgments," Discussion Paper 1998-109, Tilburg University, Center for Economic Research.
    3. Pietro Amenta & Antonio Lucadamo & Antonello D’Ambra, 2021. "Restricted Common Component and Specific Weight Analysis: A Constrained Explorative Approach for the Customer Satisfaction Evaluation," Social Indicators Research: An International and Interdisciplinary Journal for Quality-of-Life Measurement, Springer, vol. 156(2), pages 409-427, August.
    4. Elizabeth Hellier & Kirsteen Aldrich & Daniel B. Wright & Denny Daunt & Judy Edworthy, 2007. "A Multi Dimensional Analysis of Warning Signal Words," Journal of Risk Research, Taylor & Francis Journals, vol. 10(3), pages 323-338, April.
    5. Elisa Frutos-Bernal & Ángel Martín del Rey & Irene Mariñas-Collado & María Teresa Santos-Martín, 2022. "An Analysis of Travel Patterns in Barcelona Metro Using Tucker3 Decomposition," Mathematics, MDPI, vol. 10(7), pages 1-17, March.
    6. Jad Beyhum & Eric Gautier, 2020. "Factor and factor loading augmented estimators for panel regression," Working Papers hal-02957008, HAL.
    7. Yoshio Takane & Forrest Young & Jan Leeuw, 1977. "Nonmetric individual differences multidimensional scaling: An alternating least squares method with optimal scaling features," Psychometrika, Springer;The Psychometric Society, vol. 42(1), pages 7-67, March.
    8. Giuseppe Brandi & Ruggero Gramatica & Tiziana Di Matteo, 2019. "Unveil stock correlation via a new tensor-based decomposition method," Papers 1911.06126, arXiv.org, revised Apr 2020.
    9. Forrest Young & Yoshio Takane & Rostyslaw Lewyckyj, 1978. "Three notes on ALSCAL," Psychometrika, Springer;The Psychometric Society, vol. 43(3), pages 433-435, September.
    10. Herbert Marsh & Robert Boik, 1993. "Reviews," Psychometrika, Springer;The Psychometric Society, vol. 58(1), pages 145-152, March.
    11. Alwin Stegeman & Jos Berge & Lieven Lathauwer, 2006. "Sufficient conditions for uniqueness in Candecomp/Parafac and Indscal with random component matrices," Psychometrika, Springer;The Psychometric Society, vol. 71(2), pages 219-229, June.
    12. Monica Billio & Roberto Casarin & Matteo Iacopini, 2018. "Bayesian Markov Switching Tensor Regression for Time-varying Networks," Working Papers 2018:14, Department of Economics, University of Venice "Ca' Foscari".
    13. Paul Dickes & Alessio Fusco & Eric Marlier, 2010. "Structure of National Perceptions of Social Needs Across EU Countries," Social Indicators Research: An International and Interdisciplinary Journal for Quality-of-Life Measurement, Springer, vol. 95(1), pages 143-167, January.
    14. Michailidis, George & de Leeuw, Jan, 2000. "Multilevel homogeneity analysis with differential weighting," Computational Statistics & Data Analysis, Elsevier, vol. 32(3-4), pages 411-442, January.
    15. Piet Brouwer & Pieter Kroonenberg, 1991. "Some notes on the diagonalization of the extended three-mode core matrix," Journal of Classification, Springer;The Classification Society, vol. 8(1), pages 93-98, January.
    16. Krijnen, Wim P., 2006. "Convergence of the sequence of parameters generated by alternating least squares algorithms," Computational Statistics & Data Analysis, Elsevier, vol. 51(2), pages 481-489, November.
    17. Namgil Lee & Jong-Min Kim, 2018. "Block tensor train decomposition for missing data estimation," Statistical Papers, Springer, vol. 59(4), pages 1283-1305, December.
    18. Berrie Zielman & Willem Heiser, 1993. "Analysis of asymmetry by a slide-vector," Psychometrika, Springer;The Psychometric Society, vol. 58(1), pages 101-114, March.
    19. de Leeuw, Jan & Mair, Patrick, 2009. "Multidimensional Scaling Using Majorization: SMACOF in R," Journal of Statistical Software, Foundation for Open Access Statistics, vol. 31(i03).
    20. Xuan Bi & Gediminas Adomavicius & William Li & Annie Qu, 2022. "Improving Sales Forecasting Accuracy: A Tensor Factorization Approach with Demand Awareness," INFORMS Journal on Computing, INFORMS, vol. 34(3), pages 1644-1660, May.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:spr:joptap:v:197:y:2023:i:2:d:10.1007_s10957-023-02193-5. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.springer.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.