Entropy manipulation of arbitrary nonlinear mappings

Author

Fisher, John W., III ; Principe, José C.

Author_Institution

Lab. of Comput. Neuroeng., Florida Univ., Gainesville, FL, USA

fYear

1997

fDate

24-26 Sep 1997

Firstpage

14

Lastpage

23

Abstract

We discuss an unsupervised learning method which is driven by an information theoretic based criterion. The method differs from previous work in that it is extensible to a feed-forward multilayer perceptron with an arbitrary number of layers and makes no assumption about the underlying PDF of the input space. We show a simple unsupervised method by which multidimensional signals can be nonlinearly transformed onto a maximum entropy feature space resulting in statistically independent features

Keywords

feature extraction; feedforward neural nets; maximum entropy methods; multilayer perceptrons; signal processing; unsupervised learning; arbitrary nonlinear mappings; entropy manipulation; feedforward multilayer perceptron; information theoretic based criterion; maximum entropy feature space; multidimensional signals; statistically independent features; unsupervised learning method; unsupervised method; Entropy; Feature extraction; Feedforward systems; Information theory; Multilayer perceptrons; Mutual information; Neural engineering; Probability density function; Signal mapping; Unsupervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing [1997] VII. Proceedings of the 1997 IEEE Workshop

Conference_Location

Amelia Island, FL

ISSN

1089-3555

Print_ISBN

0-7803-4256-9

Type

conf

DOI

10.1109/NNSP.1997.622379

Filename

622379