pith. sign in

arxiv: 2606.22210 · v1 · pith:ST4ZIKFQnew · submitted 2026-06-20 · 💻 cs.LG · cs.AI· stat.ML

Sequential Minimal Optimization Algorithm for One-Class Support Vector Machines With Privileged Information

Andrey Lange , Dmitry Smolyakov , Evgeny Burnaev This is my paper

Pith reviewed 2026-06-26 11:57 UTC · model grok-4.3

classification 💻 cs.LG cs.AIstat.ML
keywords one-class SVMprivileged informationSMO algorithmLUPIquadratic programmingfinite convergenceanomaly detectionoptimization
0
0 comments X

The pith

An SMO algorithm for one-class SVM with privileged information converges in finite time and trains faster than interior-point solvers.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper fills a gap by extending the Learning Using Privileged Information framework to one-class SVMs, which detect anomalies using only normal training examples. It introduces a sequential minimal optimization procedure that breaks the dual quadratic program into small, solvable subproblems. The authors prove this procedure reaches the exact optimum after a finite number of steps. Experiments demonstrate that the extra privileged features, available only at training time, reshape the decision boundary in the original feature space. Direct comparisons show the new solver requires less computation than general-purpose interior-point methods on benchmark tasks.

Core claim

The authors present an SMO algorithm for the dual of the OC-SVM+ problem that selects and optimizes pairs of Lagrange multipliers at each step while holding the rest fixed. They establish that the procedure terminates at the global optimum after a finite number of iterations. Numerical tests confirm that the resulting models differ from those trained without privileged information, and that training completes more quickly than with non-sequential solvers.

What carries the argument

The SMO decomposition applied to the OC-SVM+ dual quadratic program, which reduces the full optimization to repeated two-variable subproblems that admit closed-form or simple numerical solutions.

If this is right

  • The algorithm terminates after a finite number of iterations, removing the need for arbitrary stopping tolerances.
  • Training time for OC-SVM+ models is lower than that of interior-point or other non-sequential solvers on the tested data sets.
  • The learned decision function in the original feature space changes measurably when privileged information is supplied during training.
  • The method scales to problem sizes where general-purpose quadratic solvers become impractical.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same pairwise decomposition could be applied to related one-class formulations such as support vector data description with privileged features.
  • Because privileged data is never needed at test time, the approach could improve anomaly detection pipelines that already collect extra sensor or metadata streams only during model fitting.
  • Faster, guaranteed-convergent training may make parameter sweeps over the regularization constants more feasible in practice.

Load-bearing premise

The privileged information can be folded into the one-class SVM dual problem using the standard LUPI formulation without creating numerical instabilities or needing extra tuning beyond what SVM+ and OC-SVM already require.

What would settle it

A run of the proposed SMO procedure on one of the paper's benchmark datasets that either fails to reach the same objective value as an interior-point solver or requires more wall-clock time than the non-sequential baseline.

Figures

Figures reproduced from arXiv: 2606.22210 by Andrey Lange, Dmitry Smolyakov, Evgeny Burnaev.

Figure 1
Figure 1. Figure 1: FIGURE 1 [PITH_FULL_IMAGE:figures/full_fig_p011_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIGURE 2: (2a) – OC-SVM, [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIGURE 3: Anomaly detection average precision score with [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIGURE 4: Comparison of processing times of OC-SVMs (original features and all features), OC-SVM+ via SMO with different [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗
read the original abstract

One of the powerful techniques in data modeling is accounting for features that are available at the training stage, but are not available when the trained model is used to classify or predict test data -- the Learning Using Privileged Information paradigm (LUPI). Sequential Minimal Optimization (SMO) methods have been developed for supervised Support Vector Machines (SVM), unsupervised one-class SVM, and SVM with privileged information (SVM+). The missing brick in this research has long been a one-class SVM with privileged information (OC-SVM+). In this paper, we propose an SMO algorithm for OC-SVM+ that significantly outperforms non-sequential algorithms for training the OC-SVM+ model. Its finite-time convergence is established. The experiments show how privileged information affects a descriptive domain in the space of original features. Comparative benchmark tests demonstrate that our algorithm is superior over interior point algorithms.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The manuscript proposes a Sequential Minimal Optimization (SMO) algorithm for One-Class Support Vector Machines with Privileged Information (OC-SVM+). It extends the LUPI paradigm and prior SMO work on SVM, OC-SVM, and SVM+ to this setting, claims to establish finite-time convergence of the proposed algorithm, and reports experiments showing the effect of privileged information on the descriptive domain together with outperformance relative to non-sequential and interior-point solvers.

Significance. If the finite-time convergence result is rigorously established and the empirical superiority holds under standard benchmarks, the work would complete the family of SMO solvers for SVM variants that incorporate privileged information. This could improve training efficiency for one-class models in settings where additional training-only features are available, such as anomaly detection with expert annotations.

minor comments (3)
  1. [Abstract] Abstract: the claim of finite-time convergence and experimental superiority is stated without derivation steps, precise update rules, error bounds, or any quantitative benchmark numbers, which limits immediate verifiability of the central claims.
  2. The manuscript would benefit from explicit pseudocode or update-rule equations for the SMO procedure in the OC-SVM+ dual, even if the derivation is standard.
  3. [Experiments] Experiments section: comparative results against interior-point methods should include the specific datasets, kernel choices, and numerical performance metrics (accuracy, training time, etc.) to support the superiority statement.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive assessment of our work on the SMO algorithm for OC-SVM+ and the recommendation for minor revision. No specific major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained algorithmic extension

full rationale

The paper proposes an SMO algorithm for the OC-SVM+ dual problem under the standard LUPI formulation, proves finite-time convergence, and reports empirical comparisons. No step reduces a claimed prediction or result to a fitted parameter or self-citation by construction; the convergence claim is a mathematical property of the algorithm, and performance claims are experimental outcomes. The derivation chain relies on established SVM/SMO techniques without self-referential definitions or load-bearing self-citations that collapse the central result.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are described in the abstract; the full manuscript would be required to enumerate any optimization hyperparameters, convergence assumptions, or modeling choices.

pith-pipeline@v0.9.1-grok · 5683 in / 1105 out tokens · 34365 ms · 2026-06-26T11:57:16.309471+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

94 extracted references · 1 canonical work pages

  1. [1]

    V apnik and A

    V . V apnik and A. V ashist, ‘‘A new learning paradigm: Learning using privileged information,’’Neural networks, vol. 22, no. 5-6, pp. 544–557, 2009

    2009

  2. [2]

    Platt, ‘‘Fast training of support vector machines using sequential minimal optimization,’’ inAdvances in Kernel Methods - Support V ector Learning

    J. Platt, ‘‘Fast training of support vector machines using sequential minimal optimization,’’ inAdvances in Kernel Methods - Support V ector Learning. MIT Press, January 1998. [Online]. Available: https://www.microsoft.com/en-us/research/publication/ fast-training-of-support-vector-machines-using-sequential-minimal-optimization/

    1998

  3. [3]

    S. S. Keerthi, S. K. Shevade, C. Bhattacharyya, and K. R. K. Murthy, ‘‘Improvements to Platt’s SMO algorithm for SVM classifier design,’’ Neural computation, vol. 13, no. 3, pp. 637–649, 2001

    2001

  4. [4]

    Schölkopf, J

    B. Schölkopf, J. C. Platt, J. Shawe-Taylor, A. J. Smola, and R. C. Williamson, ‘‘Estimating the support of a high-dimensional distribution,’’ Neural computation, vol. 13, no. 7, pp. 1443–1471, 2001

    2001

  5. [5]

    Pechyony and V

    D. Pechyony and V . V apnik, ‘‘Fast optimization algorithms for solving SVM+,’’Stat. Learning and Data Science, vol. 1, 2011

    2011

  6. [6]

    V apnik,Estimation of dependencies based on empirical data

    V . V apnik,Estimation of dependencies based on empirical data. Springer Science & Business Media, 2006

    2006

  7. [7]

    V apnik, A

    V . V apnik, A. V ashist, and N. Pavlovitch, ‘‘Learning using hidden informa- tion: Master-class learning,’’ inNATO workshop on Mining Massive Data sets for security. IOS Press, 2008, pp. 3–14

    2008

  8. [8]

    V apnik and R

    V . V apnik and R. Izmailov, ‘‘Knowledge transfer in SVM and neural networks,’’Annals of Mathematics and Artificial Intelligence, vol. 81, no. 1, pp. 3–19, 2017

    2017

  9. [9]

    X. Li, B. Du, C. Xu, Y . Zhang, L. Zhang, and D. Tao, ‘‘Robust learning with imperfect privileged information,’’Artificial Intelligence, vol. 282, pp. 103–246, 2020

    2020

  10. [10]

    V apnik and R

    V . V apnik and R. Izmailov, ‘‘Learning with intelligent teacher: Similarity control and knowledge transfer,’’ inInternational Symposium on Statistical Learning and Data Sciences. Springer, 2015, pp. 3–32

    2015

  11. [11]

    B. E. Boser, I. M. Guyon, and V . N. V apnik, ‘‘A training algorithm for optimal margin classifiers,’’ inProceedings of the fifth annual workshop on Computational learning theory, 1992, pp. 144–152

    1992

  12. [12]

    Cortes and V

    C. Cortes and V . V apnik, ‘‘Support-vector networks,’’Machine learning, vol. 20, no. 3, pp. 273–297, 1995

    1995

  13. [13]

    Serra-Toro, V

    C. Serra-Toro, V . J. Traver, and F. Pla, ‘‘Exploring some practical issues of SVM+: Is really privileged information that helps?’’Pattern Recognition Letters, vol. 42, pp. 40–46, 2014

    2014

  14. [14]

    Lapin, M

    M. Lapin, M. Hein, and B. Schiele, ‘‘Learning using privileged informa- tion: SVM+ and weighted SVM,’’Neural Networks, vol. 53, pp. 95–108, 2014

    2014

  15. [15]

    Lodhi, C

    H. Lodhi, C. Saunders, J. Shawe-Taylor, N. Cristianini, and C. Watkins, ‘‘Text classification using string kernels,’’Journal of machine learning research, vol. 2, no. Feb, pp. 419–444, 2002

    2002

  16. [16]

    R. I. Kondor and J. Lafferty, ‘‘Diffusion kernels on graphs and other discrete structures,’’ inProceedings of the 19th international conference on machine learning, vol. 2002, 2002, pp. 315–322

    2002

  17. [17]

    Couto, ‘‘Kernel k-means for categorical data,’’ inInternational Sympo- sium on Intelligent Data Analysis

    J. Couto, ‘‘Kernel k-means for categorical data,’’ inInternational Sympo- sium on Intelligent Data Analysis. Springer, 2005, pp. 46–56

    2005

  18. [18]

    Potdar, T

    K. Potdar, T. S. Pardawala, and C. D. Pai, ‘‘A comparative study of categorical variable encoding techniques for neural network classifiers,’’ International journal of computer applications, vol. 175, no. 4, pp. 7–9, 2017

    2017

  19. [19]

    Rodríguez, M

    P . Rodríguez, M. A. Bautista, J. Gonzalez, and S. Escalera, ‘‘Beyond one- hot encoding: Lower dimensional target embedding,’’Image and Vision Computing, vol. 75, pp. 21–31, 2018

    2018

  20. [20]

    Burnaev and D

    E. Burnaev and D. Smolyakov, ‘‘One-class SVM with privileged infor- mation and its application to malware detection,’’ in2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW). IEEE, 2016, pp. 273–280

    2016

  21. [21]

    J. Chen, X. Liu, and S. Lyu, ‘‘Boosting with side information,’’ inAsian Conference on Computer Vision. Springer, 2012, pp. 563–577

    2012

  22. [22]

    B. Liu, L. Liu, Y . Xiao, C. Liu, X. Chen, and W. Li, ‘‘AdaBoost-based transfer learning with privileged information,’’Information Sciences, vol. 593, pp. 216–232, 2022

    2022

  23. [23]

    Ribeiro, C

    B. Ribeiro, C. Silva, A. Vieira, A. Gaspar-Cunha, and J. C. das Neves, ‘‘Fi- nancial distress model prediction using svm+,’’ inThe 2010 International Joint Conference on Neural Networks (IJCNN). IEEE, 2010, pp. 1–7

    2010

  24. [24]

    Ribeiro, C

    B. Ribeiro, C. Silva, N. Chen, A. Vieira, and J. C. das Neves, ‘‘Enhanced de- fault risk models with SVM+,’’Expert Systems with Applications, vol. 39, no. 11, pp. 10 140–10 152, 2012. 17 Langeet al.: Sequential Minimal Optimization algorithm for one-class Support Vector Machines with privileged information

    2012

  25. [25]

    Liang and V

    L. Liang and V . Cherkassky, ‘‘Learning using structured data: Application to fMRI data analysis,’’ in2007 International Joint Conference on Neural Networks. IEEE, 2007, pp. 495–499

    2007

  26. [26]

    IEEE, 2008, pp

    ——, ‘‘Connection between SVM+ and multi-task learning,’’ in2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). IEEE, 2008, pp. 2048–2054

    2008

  27. [27]

    Liang, F

    L. Liang, F. Cai, and V . Cherkassky, ‘‘Predictive learning with structured (grouped) data,’’Neural Networks, vol. 22, no. 5-6, pp. 766–773, 2009

    2009

  28. [28]

    Cai and V

    F. Cai and V . Cherkassky, ‘‘SVM+ regression and multi-task learning,’’ in 2009 International Joint Conference on Neural Networks, 2009, pp. 418– 424

    2009

  29. [29]

    F. Tang, C. Xiao, F. Wang, J. Zhou, and L.-W. H. Lehman, ‘‘Retaining privileged information for multi-task learning,’’ inProceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2019, pp. 1369–1377

    2019

  30. [30]

    Y ang and I

    H. Y ang and I. Patras, ‘‘Privileged information-based conditional regres- sion forest for facial feature detection,’’ in2013 10th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG). IEEE, 2013, pp. 1–6

    2013

  31. [31]

    Sharmanska, N

    V . Sharmanska, N. Quadrianto, and C. H. Lampert, ‘‘Learning to rank using privileged information,’’ inProceedings of the IEEE international conference on computer vision, 2013, pp. 825–832

    2013

  32. [32]

    ——, ‘‘Learning to transfer privileged information,’’arXiv preprint arXiv:1410.0389, 2014

    Pith/arXiv arXiv 2014

  33. [33]

    W. Li, L. Niu, and D. Xu, ‘‘Exploiting privileged information from web data for image categorization,’’ inComputer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Pro- ceedings, Part V 13. Springer, 2014, pp. 437–452

    2014

  34. [34]

    Wang and Q

    Z. Wang and Q. Ji, ‘‘Classifier learning with hidden information,’’ in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 4969–4977

    2015

  35. [35]

    Y . Y an, F. Nie, W. Li, C. Gao, Y . Y ang, and D. Xu, ‘‘Image classification by cross-media active learning with privileged information,’’IEEE Trans- actions on Multimedia, vol. 18, no. 12, pp. 2494–2502, 2016

    2016

  36. [36]

    A. C. Rodríguez, S. D’Aronco, K. Schindler, and J. D. Wegner, ‘‘Fine- grained species recognition with privileged pooling: Better sample effi- ciency through supervised attention,’’IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 2023

    2023

  37. [37]

    Feyereisl, S

    J. Feyereisl, S. Kwak, J. Son, and B. Han, ‘‘Object localization based on structural SVM using privileged information,’’Advances in Neural Information Processing Systems, vol. 27, 2014

    2014

  38. [38]

    Lopez-Paz, L

    D. Lopez-Paz, L. Bottou, B. Schölkopf, and V . V apnik, ‘‘Unifying distilla- tion and privileged information,’’arXiv preprint arXiv:1511.03643, 2015

    Pith/arXiv arXiv 2015

  39. [39]

    L. Niu, W. Li, and D. Xu, ‘‘Exploiting privileged information from web data for action and event recognition,’’International Journal of Computer Vision, vol. 118, no. 2, pp. 130–150, 2016

    2016

  40. [40]

    Hoffman, S

    J. Hoffman, S. Gupta, and T. Darrell, ‘‘Learning with side information through modality hallucination,’’ inProceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 826–834

    2016

  41. [41]

    Sarafianos, M

    N. Sarafianos, M. Vrigkas, and I. A. Kakadiaris, ‘‘Adaptive SVM+: Learn- ing with privileged information for domain adaptation,’’ inProceedings of the IEEE International Conference on Computer Vision Workshops, 2017, pp. 2637–2644

    2017

  42. [42]

    J. Tang, Y . Tian, P . Zhang, and X. Liu, ‘‘Multiview Privileged Support V ector Machines,’’IEEE Transactions on Neural Networks and Learning Systems, vol. 29, no. 8, pp. 3463–3477, 2018

    2018

  43. [43]

    Lambert, O

    J. Lambert, O. Sener, and S. Savarese, ‘‘Deep learning under privileged information using heteroscedastic dropout,’’ inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 8886– 8895

    2018

  44. [44]

    Zhang, S

    C. Zhang, S. Sun, Y . Tian, and Z. Wang, ‘‘Structured output prediction using privileged information,’’IEEE Access, vol. 7, pp. 106 065–106 074, 2019

    2019

  45. [45]

    W. Zhu, X. Jia, and W. Li, ‘‘Privileged label enhancement with multi-label learning.’’ inIJCAI, 2020, pp. 2376–2382

    2020

  46. [46]

    Y . Li, F. Meng, J. Shi, and A. D. N. Initiative, ‘‘Learning using privileged information improves neuroimaging-based CAD of Alzheimer’s disease: a comparative study,’’Medical & biological engineering & computing, vol. 57, pp. 1605–1616, 2019

    2019

  47. [47]

    Ganaie and M

    M. Ganaie and M. Tanveer, ‘‘Ensemble deep random vector functional link network using privileged information for Alzheimer’s disease diagnosis,’’ IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2022

    2022

  48. [48]

    T. A. Shaikh, R. Ali, and M. S. Beg, ‘‘Transfer learning privileged in- formation fuels CAD diagnosis of breast cancer,’’Machine Vision and Applications, vol. 31, pp. 1–23, 2020

    2020

  49. [49]

    Zhang, L

    H. Zhang, L. Guo, J. Wang, S. Ying, and J. Shi, ‘‘Multi-View Feature Transformation Based SVM+ for Computer-Aided Diagnosis of Liver Can- cers With Ultrasound Images,’’IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 3, pp. 1512–1523, 2023

    2023

  50. [50]

    W. Lee, J. Lee, D. Kim, and B. Ham, ‘‘Learning with Privileged Informa- tion for Efficient Image Super-Resolution,’’ inComputer Vision – ECCV 2020, A. V edaldi, H. Bischof, T. Brox, and J.-M. Frahm, Eds. Cham: Springer International Publishing, 2020, pp. 465–482

    2020

  51. [51]

    Z. Wu, X. Xia, R. Wang, J. Li, J. Y u, Y . Mao, and T. Liu, ‘‘LR-SVM+: Learning using privileged information with noisy labels,’’IEEE Transac- tions on Multimedia, vol. 24, pp. 1080–1092, 2021

    2021

  52. [52]

    Z. Che, B. Liu, Y . Xiao, and H. Cai, ‘‘Twin support vector machines with privileged information,’’Information Sciences, vol. 573, pp. 141–153,

  53. [53]

    Available: https://www.sciencedirect.com/science/article/ pii/S0020025521005478

    [Online]. Available: https://www.sciencedirect.com/science/article/ pii/S0020025521005478

  54. [54]

    Y . Li, H. Sun, W. Y an, and Q. Cui, ‘‘R-CTSVM+: Robust capped L1-norm twin support vector machine with privileged information,’’Information Sciences, vol. 574, pp. 12–32, 2021

    2021

  55. [55]

    Y . Shu, Q. Li, C. Xu, S. Liu, and G. Xu, ‘‘V -SVR+: support vector regression with variational privileged information,’’IEEE Transactions on Multimedia, vol. 24, pp. 876–889, 2021

    2021

  56. [56]

    Jung and F

    B. Jung and F. D. Johansson, ‘‘Efficient learning of nonlinear prediction models with time-series privileged information,’’Advances in Neural In- formation Processing Systems, vol. 35, pp. 19 048–19 060, 2022

    2022

  57. [57]

    Feyereisl and U

    J. Feyereisl and U. Aickelin, ‘‘Privileged information for data clustering,’’ Information Sciences, vol. 194, pp. 4–23, 2012

    2012

  58. [58]

    Zhu and P

    W. Zhu and P . Zhong, ‘‘A new one-class SVM based on hidden information,’’Knowledge-Based Systems, vol. 60, pp. 35–43, 2014. [Online]. Available: https://www.sciencedirect.com/science/article/pii/ S0950705114000136

    2014

  59. [59]

    Fouad, P

    S. Fouad, P . Tino, S. Raychaudhury, and P . Schneider, ‘‘Incorporating privileged information through metric learning,’’IEEE transactions on neural networks and learning systems, vol. 24, no. 7, pp. 1086–1098, 2013

    2013

  60. [60]

    R. M. Marcacini, M. A. Domingues, E. R. Hruschka, and S. O. Rezende, ‘‘Privileged information for hierarchical document clustering: a metric learning approach,’’ in2014 22nd International Conference on Pattern Recognition. IEEE, 2014, pp. 3636–3641

    2014

  61. [61]

    Y ang, M

    X. Y ang, M. Wang, and D. Tao, ‘‘Person re-identification with metric learning using privileged information,’’IEEE Transactions on Image Pro- cessing, vol. 27, no. 2, pp. 791–805, 2017

    2017

  62. [62]

    Shekhar and L

    S. Shekhar and L. Akoglu, ‘‘Incorporating privileged information to unsu- pervised anomaly detection,’’ inJoint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, 2018, pp. 87–104

    2018

  63. [63]

    Smolyakov, N

    D. Smolyakov, N. Sviridenko, E. Burikov, and E. Burnaev, ‘‘Anomaly pattern recognition with privileged information for sensor fault detection,’’ inIAPR Workshop on Artificial Neural Networks in Pattern Recognition. Springer, 2018, pp. 320–332

    2018

  64. [64]

    T.-H. Vu, H. Jain, M. Bucher, M. Cord, and P . Pérez, ‘‘Dada: Depth- aware domain adaptation in semantic segmentation,’’ inProceedings of the IEEE/CVF International Conference on Computer Vision, 2019, pp. 7364– 7373

    2019

  65. [65]

    H. Tang, Y . Wang, and K. Jia, ‘‘Unsupervised domain adaptation via dis- tilled discriminative clustering,’’Pattern Recognition, vol. 127, Jul. 2022, art. no. 108638

    2022

  66. [66]

    Gautam, A

    C. Gautam, A. Tiwari, and M. Tanveer, ‘‘OCKELM+: kernel extreme learning machine based one-class classification using privileged informa- tion (or KOC+: Kernel ridge regression or least square SVM with zero bias based one-class classification using privileged information),’’arXiv preprint arXiv:1904.08338, 2019

    Pith/arXiv arXiv 1904

  67. [67]

    Osuna, R

    E. Osuna, R. Freund, and F. Girosi, ‘‘An improved training algorithm for support vector machines,’’ inNeural networks for signal processing VII. Proceedings of the 1997 IEEE signal processing society workshop. IEEE, 1997, pp. 276–285

    1997

  68. [68]

    Joachims, ‘‘Making large-scale SVM learning practical,’’ Technical report, Tech

    T. Joachims, ‘‘Making large-scale SVM learning practical,’’ Technical report, Tech. Rep., 1998

    1998

  69. [69]

    Platt,Sequential minimal optimization: A fast algorithm for training support vector machines

    J. Platt,Sequential minimal optimization: A fast algorithm for training support vector machines. Advances in Kernel Methods. Support V ector Learning. MIT Press, Boston, 1998. 18 Langeet al.: Sequential Minimal Optimization algorithm for one-class Support Vector Machines with privileged information

    1998

  70. [70]

    Bordes, S

    A. Bordes, S. Ertekin, J. Weston, L. Bottou, and N. Cristianini, ‘‘Fast kernel classifiers with online and active learning.’’Journal of Machine Learning Research, vol. 6, no. 9, 2005

    2005

  71. [71]

    Fan, P .-H

    R.-E. Fan, P .-H. Chen, C.-J. Lin, and T. Joachims, ‘‘Working set selection using second order information for training support vector machines.’’ Journal of machine learning research, vol. 6, no. 12, 2005

    2005

  72. [72]

    Palagi and M

    L. Palagi and M. Sciandrone, ‘‘On the convergence of a modified version of SVMlight algorithm,’’Optimization Methods and Software, vol. 20, no. 2-3, pp. 317–334, 2005. [Online]. Available: https://doi.org/10.1080/ 10556780512331318209

    2005

  73. [73]

    Glasmachers, C

    T. Glasmachers, C. Igel, K. P . Bennett, and E. Parrado-Hernández, ‘‘Maximum-gain working set selection for SVMs.’’Journal of Machine Learning Research, vol. 7, no. 7, 2006

    2006

  74. [74]

    Chang and C.-J

    C.-C. Chang and C.-J. Lin, ‘‘LIBSVM: A library for support vector machines,’’ACM Trans. Intell. Syst. Technol., vol. 2, no. 3, May 2011. [Online]. Available: https://doi.org/10.1145/1961189.1961199

    work page doi:10.1145/1961189.1961199 2011

  75. [75]

    D. M. Tax and R. P . Duin, ‘‘Support vector domain description,’’Pattern recognition letters, vol. 20, no. 11-13, pp. 1191–1199, 1999

    1999

  76. [76]

    ——, ‘‘Data domain description using support vectors.’’ inESANN, vol. 99. Citeseer, 1999, pp. 251–256

    1999

  77. [77]

    Chang, C.-P

    W.-C. Chang, C.-P . Lee, and C.-J. Lin, ‘‘A revisit to support vector data description,’’Dept. Comput. Sci., Nat. Taiwan Univ., Taipei, Taiwan, Tech. Rep, 2013

    2013

  78. [78]

    Schölkopf, R

    B. Schölkopf, R. C. Williamson, A. J. Smola, J. Shawe-Taylor, J. C. Platt et al., ‘‘Support vector method for novelty detection.’’ inNIPS, vol. 12. Citeseer, 1999, pp. 582–588

    1999

  79. [79]

    S. S. Khan and M. G. Madden, ‘‘One-class classification: taxonomy of study and review of techniques,’’The Knowledge Engineering Review, vol. 29, no. 3, pp. 345–374, 2014

    2014

  80. [80]

    C. C. Aggarwal,An introduction to outlier analysis. Springer, 2017

    2017

Showing first 80 references.