Predicting the required number of training samples

H M Kalayeh; D A Landgrebe

doi:10.1109/tpami.1983.4767459

Predicting the required number of training samples

IEEE Trans Pattern Anal Mach Intell. 1983 Jun;5(6):664-7. doi: 10.1109/tpami.1983.4767459.

Authors

H M Kalayeh¹, D A Landgrebe

Affiliation

¹ Object Recognition Systems, Inc., Princeton, NJ 08540.

PMID: 21869156
DOI: 10.1109/tpami.1983.4767459

Abstract

In this paper a criterion which measures the quality of the estimate of the covariance matrix of a multivariate normal distribution is developed. Based on this criterion, the necessary number of training samples is predicted. Experimental results which are used as a guide for determining the number of training samples are included.