Abstract
This paper establishes random multi-scale Gaussian kernels based relevance vector machine (RMGK-RVM) for regression learning problems. The mixture of multiple Gaussian kernels with different scale parameters sampled from some predefined distribution is used as the model. Under the Bayesian inference framework, RMGK-RVM can learn the whole distribution of the prediction variable. In this way, the uncertainties of input data are fully considered and the prediction accuracy of the target variable is improved for complicated data. The experimental results on one simulation data and three real-life data sets show that the proposed method performs favorably.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hofmann T, Schölkopf B, Smola AJ (2008) Kernel methods in machine learning. Ann Stat 36(3):1171–1220
Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20(3):273–297
Xu YL, Li XX, Chen DR (2018) Learning rates of regularized regression with multiple gaussian kernels for multi-task learning. IEEE Trans Neural Netw Learn Syst 29(11):5408–5418
Bach FR, Lanckriet GR, Jordan MI (2004) Multiple kernel learning, conic duality, and the SMO algorithm. In: Proceedings of the twenty-first international conference on machine learning, vol 69. ACM, New York
Gönen M, Alpaydm E (2011) Multiple kernel learning algorithms. J Mach Learn Res 12:2211–2268
Hampel FR, Ronchetti EM, Rousseeuw PJ (2011) Robust statistics: the approach based on influence functions. Wiley, New York
Tipping ME (2001) Sparse bayesian learning and the relevance vector machine. J Mach Learn Res 1:211–244
Wipf D, Nagarajan S (2007) A new view of automatic relevance determination. In: International conference on neural information processing systems, pp 1625–1632, Curran Associates Inc
Cuker F, Smale S (2011) On the mathematical foundations of learning. Bull Am Math Soc 39(1):1–49
Xin Y, Xiao GS (2009) Bayesian linear regression. Secur Ticket Control 15(1):1052–1056
Acknowledgments
This work is supported partly by First Class Discipline of Zhejiang-A (Zhejiang Gongshang University - Statistics), National Natural Science Foundation of China under grant 11571031.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Gu, Y., Dong, X., Shi, J., Kong, X. (2020). Random Multi-scale Gaussian Kernels Based Relevance Vector Machine. In: Jia, Y., Du, J., Zhang, W. (eds) Proceedings of 2019 Chinese Intelligent Systems Conference. CISC 2019. Lecture Notes in Electrical Engineering, vol 592. Springer, Singapore. https://doi.org/10.1007/978-981-32-9682-4_33
Download citation
DOI: https://doi.org/10.1007/978-981-32-9682-4_33
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-32-9681-7
Online ISBN: 978-981-32-9682-4
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)