Abstract
This article investigates optimal hypothesis selection in classification problems on the basis of a class of hypotheses distributed with respect to a posteriori probability. A method is proposed based on the concepts of relative weighted average and depth functions acting in the space of classification functions. Algorithms are developed that approximate relative depths of data and relative weighted averages and provide polynomial approximations to half-space analogues.
Similar content being viewed by others
References
R. Jornsten, Y. Vardi, and C. H. Zhang, “A robust clustering method and visualization tool based on data depth,” Statistical Data Analysis, 354–365 (2002).
J. I. Chacon, T. Duong, and M. P. Wand, “Asymptotics for general multivariate kernel density derivative estimators,” Statistics, 21, 810–837 (2011).
M. Seeger, “PAC-Bayesian generalization error bounds for Gaussian process classification,” The Journal of Machine Learning Research, 3, 237–268 (2003).
R. Herbrich, T. Graepel, and C. Campbell, “Bayes point machines,” The Journal of Machine Learning Research, 1, 247–271 (2001).
P. L. Davies and U. Gather, “Breakdown and groups,” The Annals of Statistics, 33, No. 3, 981–1027 (2005).
S. Fine, R. Gilad-Bachrach, and E. Shamir, “Query by committee, linear separation and random walks,” Theoretical Computer Science, 284, No. 1, 27–49 (2002).
S. Ben-David, N. Eiron, P. M. Long, “On the difficulty of approximately maximizing agreements,” Journal of Computer and System Sciences, 66, No. 3, 499–511 (2003).
P. J. Rousseeum and A. Struyf, “Characterizing angular symmetry and regression symmetry,” Journal of Statistical Planning and Inference, 122, 163–170 (2004).
H. Oja and D. Paindaveine, “Optimal signed-rank tests based on hyperplanes,” Journal of Statistical Planning and Inference, 135, 307–321 (2005).
C. C. Holmes and N. M. Adams, “A probabilistic nearest neighbor method for statistical pattern recognition,” Journal of the Royal Statistical Society, 64, 295–306 (2002).
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated from Kibernetika i Sistemnyi Analiz, No. 5, September–October, 2016, pp. 43–55.
Rights and permissions
About this article
Cite this article
Galkin, O.A. Algorithmic Aspects of Determining Depth Functions in a Procedure for Optimal Hypothesis Selection in Data Classification Problems. Cybern Syst Anal 52, 698–707 (2016). https://doi.org/10.1007/s10559-016-9872-8
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10559-016-9872-8