Publication de 12 numéros par an
ISSN Imprimer: 1064-2315
ISSN En ligne: 2163-9337
Indexed in
Depth-Dependent Approach to the Selection of the Optimal Hypothesis in Classification Problems
Get access
RÉSUMÉ
The research of complex approach to the selection of the optimal hypothesis in classification problems based on the class of hypotheses distributed with respect to the posterior probability is presented. The approach is based on determining the relative weighted average value for data distribution and the use of depth functions operating in the space of classification functions. Depth-dependent threshold properties of the weighted average value are studied as well as the procedure for using convex evaluative functions for the formation of posterior probabilities is improved.
-
Seeger M., PAC-Bayesian generalization error bounds for Gaussian process classification, The Journal of Machine Learning Research, 2003, 3, 237–268.
-
Herbrich R., Graepel T., Campbell C., Bayes point machines, Ibid, 2001, 1, 247–271.
-
Davies P.L., Gather U., Breakdown and groups, The Annals of Statistics, 2005, 33(3), 981–1027.
-
Zuo Y., Serfling R., General notions of statistical depth functions, Ibid, 2000, 28(2), 464–480.
-
Billor N., Abebe A., Turkmen A., Nudurupati S.V., Classification based on depth transvariations, Journal of Classification, 2008, 25(2), 253–259.
-
Fine S., Gilad-Bachrach R., Shamir E., Query by committee, linear separation and random walks, Theoretical Computer Science, 2002, 284(1), 27–49.
-
Lange T., Mosler K., Mozharovskyi P., Fast nonparametric classification based on data depth, Statistical Papers, 2014, 55, 53–67.
-
Ben-David S., Eiron N., Long P.M., On the difficulty of approximately maximizing agreements, Journal of Computer and System Sciences, 2003, 66(3), 499–511.