Abstract
In this paper we propose to combine the paradigm of multi-view semi-supervised learning with that of learning using privileged information. The combination is realized by a new method that we introduce in detail. A distinctive feature of the method is that it is classifier agnostic which contracts with most of the methods for learning using privileged information. An experimental study on a real-life problem shows that using privileged information is capable of improving multi-view semi-supervised learning.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
A view is a subset of available input variables.
- 2.
By construction for each m with \(1 < m \le M\) probability distribution \(P(\mathcal {X}_{m-1},Y)\) is a marginal distribution of probability distribution \(P(\mathcal {X}_m,Y)\).
- 3.
For example, if the prediction model is a probabilistic classifier, then the confidence value is the posterior probability of the label for an unlabeled instance.
References
Chen, T., Guestrin, C.: Xgboost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, pp. 785–794 (2016)
Chen, X., Gong, C., Ma, C., Huang, X., Yang, J:. Privileged semi-supervised learning. In: 2018 IEEE International Conference on Image Processing, ICIP 2018, Athens, Greece, 7–10 October 2018, pp. 2999–3003. IEEE (2018)
Courtnage, C., Smirnov, E.: Shapley-value data valuation for semi-supervised learning. In: Soares, C., Torgo, L. (eds.) DS 2021. LNCS (LNAI), vol. 12986, pp. 94–108. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-88942-5_8
Fouad, S., Tiño, P., Raychaudhury, S., Schneider, P.: Incorporating privileged information through metric learning. IEEE Trans. Neural Netw. Learn. Syst. 24(7), 1086–1098 (2013)
Pasunuri, R., Odom, P., Khot, T., Kersting, K., Natarajan, S.: Learning with privileged information: decision-trees and boosting. http://users.sussex.ac.uk/~nq28/beyondlabeler/PasOdoKhoKeretal16.pdf. Accessed 16 Apr 2023
Qi, Z., Tian, Y., Niu, L., Wang, B.: Semi-supervised classification with privileged information. Int. J. Mach. Learn. Cybern. 6(4), 667–676 (2015)
Smirnov, E.N., Vanderlooy, S., Sprinkhuizen-Kuyper, I.G.: Meta-typicalness approach to reliable classification. In: Proceedings of the 17th European Conference on Artificial Intelligence, ECAI 2006, vol. 141 of Frontiers in Artificial Intelligence and Applications, pp. 811–812. IOS Press (2006)
van Engelen, J.E., Hoos, H.H.: A survey on semi-supervised learning. Mach. Learn. 109(2), 373–440 (2020)
Vapnik, V., Izmailov, R.: Learning using privileged information: similarity control and knowledge transfer. J. Mach. Learn. Res. 16, 2023–2049 (2015)
Vapnik, V., Vashist, A.: A new learning paradigm: learning using privileged information. Neural Netw. 22(5–6), 544–557 (2009)
Zhao, J., Xie, X., Xin, X., Sun, S.: Multi-view learning overview: recent progress and new challenges. Inf. Fusion 38, 43–54 (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Smirnov, E., Delava, R., Diris, R., Nikolaev, N. (2023). Multi-view Semi-supervised Learning Using Privileged Information. In: Iliadis, L., Maglogiannis, I., Alonso, S., Jayne, C., Pimenidis, E. (eds) Engineering Applications of Neural Networks. EANN 2023. Communications in Computer and Information Science, vol 1826. Springer, Cham. https://doi.org/10.1007/978-3-031-34204-2_13
Download citation
DOI: https://doi.org/10.1007/978-3-031-34204-2_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-34203-5
Online ISBN: 978-3-031-34204-2
eBook Packages: Computer ScienceComputer Science (R0)