Anti-correlation: A Diversity Promoting Mechanisms
in Ensemble Learning
R. McKay, and H. Abbass. The Australian Journal of Intelligent Information
Processing Systems, 7 (3/4):
139--149(2001)
Abstract
Anticorrelation has been used in training neural
network ensembles. Negative correlation learning (NCL)
is the state of the art anticorrelation measure. We
present an alternative anticorrelation measure,
RTQRTNCL, which shows significant improvements on our
test examples for both artificial neural networks (ANN)
and genetic programming (GP) learning machines. We
analyse the behaviour of the negative correlation
measure and derive a theoretical explanation of the
improved performance of RTQRTNCL in larger ensembles.
%0 Journal Article
%1 McKay:2001:AJIIPS_2
%A McKay, R. I. (Bob)
%A Abbass, H. A.
%D 2001
%J The Australian Journal of Intelligent Information
Processing Systems
%K Anticorrelation, Artificial Ensemble Networks, Neural algorithms, committee diversity fitness genetic learning, programming, sharing,
%N 3/4
%P 139--149
%T Anti-correlation: A Diversity Promoting Mechanisms
in Ensemble Learning
%U http://sc.snu.ac.kr/PAPERS/AJIIPS_anticorr.pdf
%V 7
%X Anticorrelation has been used in training neural
network ensembles. Negative correlation learning (NCL)
is the state of the art anticorrelation measure. We
present an alternative anticorrelation measure,
RTQRTNCL, which shows significant improvements on our
test examples for both artificial neural networks (ANN)
and genetic programming (GP) learning machines. We
analyse the behaviour of the negative correlation
measure and derive a theoretical explanation of the
improved performance of RTQRTNCL in larger ensembles.
@article{McKay:2001:AJIIPS_2,
abstract = {Anticorrelation has been used in training neural
network ensembles. Negative correlation learning (NCL)
is the state of the art anticorrelation measure. We
present an alternative anticorrelation measure,
RTQRTNCL, which shows significant improvements on our
test examples for both artificial neural networks (ANN)
and genetic programming (GP) learning machines. We
analyse the behaviour of the negative correlation
measure and derive a theoretical explanation of the
improved performance of RTQRTNCL in larger ensembles.},
added-at = {2008-06-19T17:35:00.000+0200},
author = {McKay, R. I. (Bob) and Abbass, H. A.},
biburl = {https://www.bibsonomy.org/bibtex/2fae2bfea4cea86d15a98d93894080457/brazovayeye},
interhash = {589c849df8460350ec5c480b74124935},
intrahash = {fae2bfea4cea86d15a98d93894080457},
journal = {The Australian Journal of Intelligent Information
Processing Systems},
keywords = {Anticorrelation, Artificial Ensemble Networks, Neural algorithms, committee diversity fitness genetic learning, programming, sharing,},
number = {3/4},
pages = {139--149},
timestamp = {2008-06-19T17:46:40.000+0200},
title = {Anti-correlation: {A} Diversity Promoting Mechanisms
in Ensemble Learning},
url = {http://sc.snu.ac.kr/PAPERS/AJIIPS_anticorr.pdf},
volume = 7,
year = 2001
}