Abstract
The generalized pinball loss function was introduced to improve the noise sensitivity and random instability of the original support vector machine (SVM). However, because the generalized pinball loss function is not smooth, SVM that uses it (GP-loss-SVM) is non-differentiable. From this issue, it makes the several practical algorithms that cannot be used to find the solution. As a result, GP-loss-SVM performs less effectively in practice. In order to solve this problem and improve GP-loss-SVM, in this paper, we construct a new smooth approximation function of the generalized pinball loss function into the SVM. As a result, a smooth SVM with the generalized pinball loss function is obtained. Furthermore, we use an effective method, quasi-Newton–Armijo, to solve our model. Moreover, we prove that our new loss function can estimate the generalized pinball loss function. Finally, we conduct a thorough experimental investigation employing a variety of machine learning benchmark datasets. The experiment results on binary datasets show that the proposed method outperforms the original SVM method by up to 1.59% for linear kernel and 1.45% for nonlinear kernel; the experiment results on multi-class datasets show that the proposed method outperforms the baseline model by up to 1.43% for linear kernels. We also evaluate the statistical significance of the performance values attained for the various models using the Friedman test. The results demonstrate that the average rank of the proposed method is better compared to baseline models.
Similar content being viewed by others
References
Cortes C, Vapnik VN (1995) Support vector networks. Mach Learn 20(3):273–297
Zhai Z, Gu B, Li X, Huang H (2020) Safe sample screening for robust support vector machine. AAAI
Liang Z, Li Y (2009) Incremental support vector machine learning in the primal and applications. Neurocomputing 72:2249–2258
Chen WJ, Li CN, Shao YH, Zhang J, Deng NY (2018) Robust l1-norm multi-weight vector projection support vector machine with efficient algorithm. Neurocomputing 315:345–361
Joachims T (1998) Text categorization with suport vector machines: learning with many relevant features. Mach Learn 1398:137–142
Yin H, Jiao X, Chai Y, Fang B (2015) Scene classification based on single-layer sae and svm. Expert Syst Appl 42(7):3368–3380
Dniz O, Castrilln M, Hernndez M (2003) Face recognition using independent component analysis and support vector machines. Pattern Recogn Lett 24(13):2153–2157
Burges CJC (1998) A tutorial on support vector machines for pattern recognition. Data Min Knowl Disc 2(2):121–167
Huang X, Shi L, Suykens JA (2014) Support vector machine classifier with pinball loss. IEEE Trans Pattern Analys Mach Intell 36(5):984–97
Gupta U, Gupta D (2019) An improved regularization based Lagrangian asymmetric -twin support vector regression using pinball loss function. Appl Intell 49(10):3606–3627
Gupta D, Gupta U (2021) On robust asymmetric Lagrangian -twin support vector regression using pinball loss function. Appl Soft Comput 102:107099
Feng R, Xu Y (2022) Support matrix machine with pinball loss for classification. Neural Comput Appl 34(21):18643–18661
Reshma R, Khemchandani R, Pal A, Chandra S (2018) Generalized pinball loss svms. Neurocomputing 36(5):322
Huang X, Shi L, Johan AK (2015) SuykensSequential minimal optimization for SVM with pinball loss. Neurocomputing 149:1596–1603. https://doi.org/10.1016/j.neucom.2014.08.033
Wang J, Wonka P, Ye J (2014) Scaling SVM and least absolute deviations via exact data reduction. In: Proceedings of the 31st International Conference on Machine Learning, PMLR. 32(2), 523-531
Vaněk J, Michálek J, Psutka J (2017) A GPU-architecture optimized hierarchical decomposition algorithm for support vector machine training. IEEE Trans Parallel Distrib Syst 28(12):3330–3343. https://doi.org/10.1109/TPDS.2017.2731764
Shalev-Shwartz S, Singer Y, Srebro N (2011) Pegasos: primal estimated sub-gradient solver for SVM. Math Program 127:3–30. https://doi.org/10.1007/s10107-010-0420-4
Wang Z, Shao YH, Bai L, Li C, Liu L, Denge N (2018) Insensitive stochastic gradient twin support vector machines for large scale problems. Inf Sci 462(1):14–31
Panup W, Wangkeeree R (2021) Stochastic subgradient for large-scale support vector machine using the generalized pinball loss function. Symmetry 13(9):1652. https://doi.org/10.3390/sym13091652
Arun Kumar M, Gopal M (2008) Application of smoothing technique on twin support vector machines. Pattern Recognit Lett 29(13):1842–1848. https://doi.org/10.1016/j.patrec.2008.05.016
Li K, Lv Z (2021) Smooth twin bounded support vector machine with pinball loss. Appl Intell 51:5489–5505
Xiong J, Hu T, Li G, Peng H (2006) A comparative study of three smooth SVM classifiers. In: 2006 6th World Congress on Intelligent Control and Automation. 5962–5966, doi: https://doi.org/10.1109/WCICA.2006.1714223
Qin C, Liu S (2012) Fuzzy smooth support vector machine with different smooth functions. J Syst Eng Electron 23(3):460–466. https://doi.org/10.1109/JSEE.2012.00058
Sharma S, Rastogi R, Chandra S (2021) Large-scale twin parametric support vector machine using pinball loss function. IEEE Trans Syst, Man, and Cyberne: Syst 51(2):987–1003
Liang ZL, Zhang L (2021) Support vector machines with the \(\epsilon\)-insensitive pinball loss function for uncertain data classification. Neurocomputing 457:117–127. https://doi.org/10.1016/j.neucom.2021.06.044
Lee Y, Mangasarian OL (2001) SSVM: a smooth support vector machine for classification. Comput Optim Appl 20(1):5–22
Ding S, Huang H, Xu X, Wang J (2014) Polynomial smooth twin support vector machines. Appl Math, Inf Sci 8:2063–2071
Tanveer M (2017) Smooth twin support vector machines via unconstrained convex minimization. Filomat 31:2195–2210
Liu MZ, Shao YH, Li CN, Chen WJ (2021) Smooth pinball loss nonparallel support vector machine for robust classification. Appl Soft Comput 98:106840
Ratiphaphongthon W, Panup W, Wangkeeree R (2022) An improved technique for pneumonia infected patients image recognition based on combination algorithm of smooth generalized pinball SVM and variational autoencoders. IEEE Access 10:107431–107445. https://doi.org/10.1109/ACCESS.2022.3212535
Panup W, Ratipapongton W, Wangkeeree R (2022) A novel twin support vector machine with generalized pinball loss function for pattern classification. Symmetry 14(2):289
Ralph Tyrell Rockafellar, Roger J-B Wets.: Variational analysis. Springer. (1998)
Proximite et dualtité dans un espace Hilbertien (1965) Jean J. Moreau. Búlletin de la Société Mathematique de Francé. 93:273–299
Boyd S, Vandenberghe L (2004) Convex optimization. Cambridge University Press. http://stanford.edu/boyd/cvxbook/
Nocedal J, Wright SJ (1999) Numerical optimization, 2nd edn. Springer-Verlag, New York, NY
Bache K, Lichman M (2013) UCI machine learning repository. [Online]. Available: http://archive.ics.uci.edu/ml (1990):92
Pedregosa et al (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825–2830
Hsu CW, Chang CC, Lin CJ (2003) A practical guide to support vector classification. Nat. Taiwan Univ, Taipei, Taiwan, pp 1–12
Garcia S, Fernandez A, Luengo J, Herrera F (2010) Advanced non- parametric tests for multiple comparisons in the design of experiments in computational intelligence and data mining experimental analysis of power. Inf Sci 180:2044–2064
Acknowledgements
This work was supported in part by the Office of National Higher Education Science Research and Innovation Policy and Naresuan University, Thailand, with Grant Number B05F640180.
Funding
This research was funded by Office of National Higher Education Science Research and Innovation Policy and Naresuan University, Thailand, with Grant Number B05F640180.
Author information
Authors and Affiliations
Contributions
Authors’ contributions Dawrawee Makmuang constructed the model, analyzed its properties, and wrote the code. WR tested the model and performed the numerical experiments. RW reviewed and edited the original draft. The published version of the manuscript has been read and approved by all authors.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no conflicts of interest to declare that are relevant to the content of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
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
Makmuang, D., Ratiphaphongthon, W. & Wangkeeree, R. Smooth support vector machine with generalized pinball loss for Pattern Classification. J Supercomput 79, 11684–11706 (2023). https://doi.org/10.1007/s11227-023-05082-w
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-023-05082-w