Accurate Bayesian Data Classification without Hyperparameter Cross-validation

DOI10.1007/S00357-019-09316-6MaRDI QIDQ6295796arXivFDO

Authors Mansoor Sheikh, A. C. C. Coolen

Publication date 28 December 2017

Published in arXiv

Full work available at URL https://doi.org/10.1007/S00357-019-09316-6

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30)

arXiv classification

cs.LG math.ST stat.ME stat.TH

Abstract: We extend the standard Bayesian multivariate Gaussian generative data classifier by considering a generalization of the conjugate, normal-Wishart prior distribution and by deriving the hyperparameters analytically via evidence maximization. The behaviour of the optimal hyperparameters is explored in the high-dimensional data regime. The classification accuracy of the resulting generalized model is competitive with state-of-the art Bayesian discriminant analysis methods, but without the usual computational burden of cross-validation.

This page was built for publication: Accurate Bayesian Data Classification without Hyperparameter Cross-validation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6295796)