Skip to main content
Springer Nature Link
Log in

Speeding Up Budgeted Stochastic Gradient Descent SVM Training with Precomputed Golden Section Search

  • Conference paper
  • First Online:
Machine Learning, Optimization, and Data Science (LOD 2018)

Part of the book series: Lecture Notes in Computer Science ((LNISA,volume 11331))

  • 2191 Accesses

Abstract

Limiting the model size of a kernel support vector machine to a pre-defined budget is a well-established technique that allows to scale SVM learning and prediction to large-scale data. Its core addition to simple stochastic gradient training is budget maintenance through merging of support vectors. This requires solving an inner optimization problem with an iterative method many times per gradient step. In this paper we replace the iterative procedure with a fast lookup. We manage to reduce the merging time by up to \(65\%\) and the total training time by \(44\%\) without any loss of accuracy.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Similar content being viewed by others

Notes

  1. 1.

    https://www.ini.rub.de/the_institute/people/tobias-glasmachers/#software .

  2. 2.

    Note that with increasing number of passes (or epochs) the standard deviation does not tend to zero since the training problem is non-convex due to the budget constraint.

References

  1. Bottou, L., Lin, C.J.: Support Vector Machine Solvers, pp. 1–28. MIT Press, Cambridge (2007)

    Google Scholar 

  2. Bottou, L.: Large-scale machine learning with stochastic gradient descent. In: Lechevallier, Y., Saporta, G. (eds.) COMPSTAT 2010, pp. 177–186. Physica-Verlag, Heidelberg (2010). https://doi.org/10.1007/978-3-7908-2604-3_16

    Chapter  Google Scholar 

  3. Burges, C.J.: Simplified support vector decision rules, pp. 71–77. Morgan Kaufmann (1996)

    Google Scholar 

  4. Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines. ACM Trans. Intell. Syst. Technol. 2, 27 (2011)

    Article  Google Scholar 

  5. Cortes, C., Vapnik, V.: Support-vector networks. Mach. Learn. 20(3), 273–297 (1995)

    MATH  Google Scholar 

  6. Cui, J., Li, Z., Lv, R., Xu, X., Gao, J.: The application of support vector machine in pattern recognition. IEEE Trans. Control Autom. (2007)

    Google Scholar 

  7. Fine, S., Scheinberg, K.: Efficient SVM training using low-rank kernel representations. J. Mach. Learn. Res. 2(Dec), 243–264 (2001)

    MATH  Google Scholar 

  8. Graf, H.P., Cosatto, E., Bottou, L., Dourdanovic, I., Vapnik, V.: Parallel support vector machines: the cascade SVM. In: NIPS (2005)

    Google Scholar 

  9. Hare, S., et al.: Struck: structured output tracking with kernels. IEEE Trans. Pattern Anal. Mach. Intell. 38(10), 2096–2109 (2016)

    Article  Google Scholar 

  10. Hsieh, C.J., Chang, K.W., Lin, C.J., Keerthi, S.S., Sundararajan, S.: A dual coordinate descent method for large-scale linear SVM. In: ICML (2008)

    Google Scholar 

  11. Hsieh, C.J., Si, S., Dhillon, I.: A divide-and-conquer solver for kernel support vector machines. In: International Conference on Machine Learning (ICML), pp. 566–574 (2014)

    Google Scholar 

  12. Joachims, T.: Text categorization with support vector machines: learning with many relevant features. In: Nédellec, C., Rouveirol, C. (eds.) ECML 1998. LNCS, vol. 1398, pp. 137–142. Springer, Heidelberg (1998). https://doi.org/10.1007/BFb0026683

    Chapter  Google Scholar 

  13. Joachims, T.: Making large-scale SVM learning practical. In: Advances in Kernel Methods - Support Vector Learning. MIT Press, Cambridge (1999)

    Google Scholar 

  14. Ladicky, L., Torr, P.: Locally linear support vector machines. In: International Conference on Machine Learning (ICML), pp. 985–992 (2011)

    Google Scholar 

  15. Lewis, D.P., Jebara, T., Noble, W.S.: Support vector machine learning from heterogeneous data: an empirical analysis using protein sequence and structure. Bioinformatics 22(22), 2753–2760 (2006)

    Article  Google Scholar 

  16. Lin, G., Shen, C., Shi, Q., van den Hengel, A., Suter, D.: Fast supervised hashing with decision trees for high-dimensional data. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2014)

    Google Scholar 

  17. Mohri, M., Rostamizadeh, A., Talwalkar, A.: Foundations of Machine Learning. MIT Press, Cambridge (2012)

    MATH  Google Scholar 

  18. Nguyen, D., Ho, T.: An efficient method for simplifying support vector machines. In: Proceedings of the 22nd ICML, pp. 617–624 (2005)

    Google Scholar 

  19. Noble, W.S.: Support vector machine applications in computational biology. In: Schölkopf, B., Tsuda, K., Vert, J.P. (eds.) Kernel Methods in Computational Biology. MIT Press, Cambridge (2004)

    Google Scholar 

  20. Rahimi, A., Recht, B.: Random features for large-scale kernel machines. NIPS 3(4) (2007)

    Google Scholar 

  21. Schölkopf, B., et al.: Input space versus feature space in kernel-based methods. IEEE Trans. Neural Netw. 10(5), 1000–1017 (1999)

    Article  Google Scholar 

  22. Shalev-Shwartz, S., Singer, Y., Srebro, N., Cotter, A.: Pegasos: primal estimated sub-GrAdient SOlver for SVM. Math. Program. 127(1), 3–30 (2011)

    Article  MathSciNet  MATH  Google Scholar 

  23. Steinwart, I.: Sparseness of support vector machines. J. Mach. Learn. Res. 4(Nov), 1071–1105 (2003)

    MathSciNet  MATH  Google Scholar 

  24. Vapnik, V.: The Nature of Statistical Learning Theory. Springer, New York (1995). https://doi.org/10.1007/978-1-4757-2440-0

    Book  MATH  Google Scholar 

  25. Wang, Z., Crammer, K., Vucetic, S.: Breaking the curse of kernelization: budgeted stochastic gradient descent for large-scale SVM training. J. Mach. Learn. Res. 13, 3103–3131 (2012)

    MathSciNet  MATH  Google Scholar 

  26. Wen, Z., Shi, J., He, B., Li, Q., Chen, J.: Thunder-SVM (2017). https://github.com/zeyiwen/thundersvm

  27. Mu, Y., Hua, G., Fan, W., Chang, S.F.: Hash-SVM: scalable kernel machines for large-scale visual classification. In: IEEE Conference on Computer Vision and Pattern Recognition (2014)

    Google Scholar 

  28. Yu, J., Xue, A., Redei, E., Bagheri, N.: A support vector machine model provides an accurate transcript-level-based diagnostic for major depressive disorder. Transl. Psychiatry 6(10), e931 (2016). https://doi.org/10.1038/tp.2016.198

    Article  Google Scholar 

  29. Zanni, L., Serafini, T., Zanghirati, G.: Parallel software for training large scale support vector machines on multiprocessor systems. J. Mach. Learn. Res. 7, 1467–1492 (2006)

    MathSciNet  MATH  Google Scholar 

  30. Zhang, H., Berg, A.C., Maire, M., Malik, J.: SVM-KNN: discriminative nearest neighbor classification for visual category recognition. In: Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 2126–2136. IEEE (2006)

    Google Scholar 

  31. Zhang, K., Lan, L., Wang, Z., Moerchen, F.: Scaling up kernel SVM on limited resources: a low-rank linearization approach. In: AISTATS (2012)

    Google Scholar 

  32. Zhang, T.: Solving large scale linear prediction problems using stochastic gradient descent. In: International Conference on Machine Learning (2004)

    Google Scholar 

  33. Zhu, Z.A., Chen, W., Wang, G., Zhu, C., Chen, Z.: P-packSVM: parallel primal gradient descent kernel SVM. In: IEEE International Conference on Data Mining (2009)

    Google Scholar 

Download references

Acknowledgments

We acknowledge support by the Deutsche Forschungsgemeinschaft (DFG) through grant GL 839/3-1.

Author information

Authors and Affiliations

  1. Institute for Neural Computation, Ruhr University Bochum, 44801, Bochum, NRW, Germany

    Tobias Glasmachers & Sahar Qaadan

Authors
  1. Tobias Glasmachers
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sahar Qaadan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Tobias Glasmachers .

Editor information

Editors and Affiliations

  1. University of Catania, Catania, Italy and University of Reading, Reading, UK

    Giuseppe Nicosia

  2. University of Florida, Gainesville, FL, USA

    Panos Pardalos

  3. University of Catania, Catania, Italy

    Giovanni Giuffrida

  4. Harvard University, Cambridge, MA, USA

    Renato Umeton

  5. IBM, Tivoli Research Lab, Rome, Italy

    Vincenzo Sciacca

Rights and permissions

Reprints and permissions

Copyright information

© 2019 Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Glasmachers, T., Qaadan, S. (2019). Speeding Up Budgeted Stochastic Gradient Descent SVM Training with Precomputed Golden Section Search. In: Nicosia, G., Pardalos, P., Giuffrida, G., Umeton, R., Sciacca, V. (eds) Machine Learning, Optimization, and Data Science. LOD 2018. Lecture Notes in Computer Science(), vol 11331. Springer, Cham. https://doi.org/10.1007/978-3-030-13709-0_28

Download citation

  • DOI: https://doi.org/10.1007/978-3-030-13709-0_28

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-13708-3

  • Online ISBN: 978-3-030-13709-0

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics