DocumentCode :
1021408
Title :
Entropic Volterra classifier (EVC) for use in data classification
Author :
Lasenby, Joan ; Fitzgerald, William J.
Author_Institution :
Dept. of Eng., Cambridge Univ.
Volume :
30
Issue :
1
fYear :
1994
fDate :
1/6/1994 12:00:00 AM
Firstpage :
53
Lastpage :
54
Abstract :
The authors consider the use of a novel form of connectionist model as a pattern classifier. The major difference in this approach is the introduction of an entropic regularising term into the basic least-squares problem. The network weights are found by optimising this regularised cost function over the training data
Keywords :
least squares approximations; neural nets; pattern recognition; connectionist model; data classification; entropic Volterra classifier; entropic regularising term; least-squares problem; network weights; pattern classifier; regularised cost function; training data;
fLanguage :
English
Journal_Title :
Electronics Letters
Publisher :
iet
ISSN :
0013-5194
Type :
jour
DOI :
10.1049/el:19940005
Filename :
260598
Link To Document :
بازگشت