IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2411.19285.html
   My bibliography  Save this paper

BPQP: A Differentiable Convex Optimization Framework for Efficient End-to-End Learning

Author

Listed:
  • Jianming Pan
  • Zeqi Ye
  • Xiao Yang
  • Xu Yang
  • Weiqing Liu
  • Lewen Wang
  • Jiang Bian

Abstract

Data-driven decision-making processes increasingly utilize end-to-end learnable deep neural networks to render final decisions. Sometimes, the output of the forward functions in certain layers is determined by the solutions to mathematical optimization problems, leading to the emergence of differentiable optimization layers that permit gradient back-propagation. However, real-world scenarios often involve large-scale datasets and numerous constraints, presenting significant challenges. Current methods for differentiating optimization problems typically rely on implicit differentiation, which necessitates costly computations on the Jacobian matrices, resulting in low efficiency. In this paper, we introduce BPQP, a differentiable convex optimization framework designed for efficient end-to-end learning. To enhance efficiency, we reformulate the backward pass as a simplified and decoupled quadratic programming problem by leveraging the structural properties of the KKT matrix. This reformulation enables the use of first-order optimization algorithms in calculating the backward pass gradients, allowing our framework to potentially utilize any state-of-the-art solver. As solver technologies evolve, BPQP can continuously adapt and improve its efficiency. Extensive experiments on both simulated and real-world datasets demonstrate that BPQP achieves a significant improvement in efficiency--typically an order of magnitude faster in overall execution time compared to other differentiable optimization layers. Our results not only highlight the efficiency gains of BPQP but also underscore its superiority over differentiable optimization layer baselines.

Suggested Citation

  • Jianming Pan & Zeqi Ye & Xiao Yang & Xu Yang & Weiqing Liu & Lewen Wang & Jiang Bian, 2024. "BPQP: A Differentiable Convex Optimization Framework for Efficient End-to-End Learning," Papers 2411.19285, arXiv.org, revised Dec 2024.
  • Handle: RePEc:arx:papers:2411.19285
    as

    Download full text from publisher

    File URL: http://arxiv.org/pdf/2411.19285
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Andrew Butler & Roy H. Kwon, 2023. "Efficient differentiable quadratic programming layers: an ADMM approach," Computational Optimization and Applications, Springer, vol. 84(2), pages 449-476, March.
    2. Adam N. Elmachtoub & Paul Grigas, 2022. "Smart “Predict, then Optimize”," Management Science, INFORMS, vol. 68(1), pages 9-26, January.
    3. Brendan O’Donoghue & Eric Chu & Neal Parikh & Stephen Boyd, 2016. "Conic Optimization via Operator Splitting and Homogeneous Self-Dual Embedding," Journal of Optimization Theory and Applications, Springer, vol. 169(3), pages 1042-1068, June.
    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. Andrew Butler & Roy H. Kwon, 2023. "Efficient differentiable quadratic programming layers: an ADMM approach," Computational Optimization and Applications, Springer, vol. 84(2), pages 449-476, March.
    2. Enzo Busseti, 2019. "Derivative of a Conic Problem with a Unique Solution," Papers 1903.05753, arXiv.org, revised Mar 2019.
    3. da Costa, B. Freitas Paulo & Pesenti, Silvana M. & Targino, Rodrigo S., 2023. "Risk budgeting portfolios from simulations," European Journal of Operational Research, Elsevier, vol. 311(3), pages 1040-1056.
    4. Leo Liberti, 2020. "Distance geometry and data science," TOP: An Official Journal of the Spanish Society of Statistics and Operations Research, Springer;Sociedad de Estadística e Investigación Operativa, vol. 28(2), pages 271-339, July.
    5. Richard Spady & Sami Stouli, 2020. "Gaussian Transforms Modeling and the Estimation of Distributional Regression Functions," Papers 2011.06416, arXiv.org.
    6. Johannes Friedrich & Pengcheng Zhou & Liam Paninski, 2017. "Fast online deconvolution of calcium imaging data," PLOS Computational Biology, Public Library of Science, vol. 13(3), pages 1-26, March.
    7. Florian Schwendinger & Bettina Grün & Kurt Hornik, 2021. "A comparison of optimization solvers for log binomial regression including conic programming," Computational Statistics, Springer, vol. 36(3), pages 1721-1754, September.
    8. Andrew Butler & Roy Kwon, 2021. "Efficient differentiable quadratic programming layers: an ADMM approach," Papers 2112.07464, arXiv.org.
    9. Estey, Clayton, 2024. "Robust Bellman State Prediction with Learning and Model Preferences," OSF Preprints 75fc9, Center for Open Science.
    10. Jinhak Kim & Mohit Tawarmalani & Jean-Philippe P. Richard, 2019. "Convexification of Permutation-Invariant Sets," Purdue University Economics Working Papers 1315, Purdue University, Department of Economics.
    11. Sun, Qinghe & Chen, Li & Meng, Qiang, 2022. "Evaluating port efficiency dynamics: A risk-based approach," Transportation Research Part B: Methodological, Elsevier, vol. 166(C), pages 333-347.
    12. Chen, Ying & Koch, Thorsten & Zakiyeva, Nazgul & Zhu, Bangzhu, 2020. "Modeling and forecasting the dynamics of the natural gas transmission network in Germany with the demand and supply balance constraint," Applied Energy, Elsevier, vol. 278(C).
    13. Stephen Boyd & Kasper Johansson & Ronald Kahn & Philipp Schiele & Thomas Schmelzer, 2024. "Markowitz Portfolio Construction at Seventy," Papers 2401.05080, arXiv.org.
    14. Benoît Legat & Oscar Dowson & Joaquim Dias Garcia & Miles Lubin, 2022. "MathOptInterface: A Data Structure for Mathematical Optimization Problems," INFORMS Journal on Computing, INFORMS, vol. 34(2), pages 672-689, March.
    15. Run Chen & Andrew L. Liu, 2021. "A distributed algorithm for high-dimension convex quadratically constrained quadratic programs," Computational Optimization and Applications, Springer, vol. 80(3), pages 781-830, December.
    16. Nikitas Rontsis & Paul Goulart & Yuji Nakatsukasa, 2022. "Efficient Semidefinite Programming with Approximate ADMM," Journal of Optimization Theory and Applications, Springer, vol. 192(1), pages 292-320, January.
    17. Chris Coey & Lea Kapelevich & Juan Pablo Vielma, 2022. "Solving Natural Conic Formulations with Hypatia.jl," INFORMS Journal on Computing, INFORMS, vol. 34(5), pages 2686-2699, September.
    18. Pietro D’Alessandro & Manlio Gaudioso & Giovanni Giallombardo & Giovanna Miglionico, 2024. "The Descent–Ascent Algorithm for DC Programming," INFORMS Journal on Computing, INFORMS, vol. 36(2), pages 657-671, March.
    19. Nicholas Moehle & Mykel J. Kochenderfer & Stephen Boyd & Andrew Ang, 2021. "Tax-Aware Portfolio Construction via Convex Optimization," Journal of Optimization Theory and Applications, Springer, vol. 189(2), pages 364-383, May.
    20. Ali Shirali & Rediet Abebe & Moritz Hardt, 2024. "Allocation Requires Prediction Only if Inequality Is Low," Papers 2406.13882, arXiv.org.

    More about this item

    Statistics

    Access and download statistics

    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:arx:papers:2411.19285. 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: arXiv administrators (email available below). General contact details of provider: http://arxiv.org/ .

    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.