Learning from examples with quadratic mutual information

Author

Xu, Dongxin ; Principe, Jose C.

Author_Institution

Comput. NeuroEng. Lab., Florida Univ., Gainesville, FL, USA

fYear

1998

fDate

31 Aug-2 Sep 1998

Firstpage

155

Lastpage

164

Abstract

Discusses an algorithm to train nonlinear mappers with information theoretic criteria (entropy or mutual information) directly from a training set. The method is based on a Parzen window estimator and uses Renyi´s quadratic definition of entropy and a distance measure based on the Cauchy-Schwartz inequality. We apply the algorithm to the difficult problem of vehicle pose estimation in synthetic aperture radar (SAR) with very good results

Keywords

covariance matrices; entropy; estimation theory; image processing; learning by example; multilayer perceptrons; nonparametric statistics; synthetic aperture radar; Cauchy-Schwartz inequality; Parzen window estimator; distance measure; entropy; information theoretic criteria; mutual information; nonlinear mappers; quadratic mutual information; synthetic aperture radar; vehicle pose estimation; Entropy; Information analysis; Information processing; Information theory; Laboratories; Mutual information; Neural engineering; Pattern recognition; Synthetic aperture radar; Vehicles;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing VIII, 1998. Proceedings of the 1998 IEEE Signal Processing Society Workshop

Conference_Location

Cambridge

ISSN

1089-3555

Print_ISBN

0-7803-5060-X

Type

conf

DOI

10.1109/NNSP.1998.710645

Filename

710645