Abstract
Many applications of machine learning involve sparse high-dimensional data, where the number of input features is (much) larger than the number of data samples, d ≫ n. Predictive modeling of such data is very ill-posed and prone to overfitting. Several recent studies for modeling high-dimensional data employ new learning methodology called Learning through Contradictions or Universum Learning due to Vapnik (1998,2006). This method incorporates a priori knowledge about application data, in the form of additional Universum samples, into the learning process. This paper investigates generalization properties of the Universum-SVM and how they are related to characteristics of the data. We describe practical conditions for evaluating the effectiveness of Random Averaging Universum.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ahn, J., Marron, J.S.: The direction of maximal data piling in high dimensional space. Technical Report, University of North Carolina at Chapel Hill (2005)
Cherkassky, V., Mulier, F.: Learning from Data Concepts: Theory and Methods, 2nd edn. Wiley, NY (2007)
Schölkopf, B., Smola, A.: Learning with Kernels. MIT Press, Cambridge (2002)
Vapnik, V.N.: Statistical Learning Theory. Wiley, NY (1998)
Vapnik, V.N.: Estimation of Dependencies Based on Empirical Data. In: Empirical Inference Science: Afterword of 2006. Springer, Heidelberg (2006)
Weston, J., Collobert, R., Sinz, F., Bottou, L.: V.Vapnik, Inference with the Universum. In: Proc. ICML (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Cherkassky, V., Dai, W. (2009). Empirical Study of the Universum SVM Learning for High-Dimensional Data. In: Alippi, C., Polycarpou, M., Panayiotou, C., Ellinas, G. (eds) Artificial Neural Networks – ICANN 2009. ICANN 2009. Lecture Notes in Computer Science, vol 5768. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-04274-4_96
Download citation
DOI: https://doi.org/10.1007/978-3-642-04274-4_96
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-04273-7
Online ISBN: 978-3-642-04274-4
eBook Packages: Computer ScienceComputer Science (R0)