This paper applies the evolution of GP teams to
different classification and regression problems and
compares different methods for combining the outputs of
the team programs. These include hybrid approaches
where (1) a neural network is used to optimize the
weights of programs in a team for a common decision and
(2) a real numbered vector (the representation of
evolution strategies) of weights is evolved with each
term in parallel. The cooperative team approach results
in an improved training and generalization performance
compared to the standard GP method. The higher
computational overhead of team evolution is
counteracted by using a fast variant of linear GP. In
particular, the processing time of linear genetic
programs is reduced significantly by removing intron
code before program execution.
%0 Journal Article
%1 brameier:2001:GPEM
%A Brameier, Markus
%A Banzhaf, Wolfgang
%D 2001
%J Genetic Programming and Evolvable Machines
%K algorithms, combination evolution genetic linear multiple of predictors, programming programming, teams,
%N 4
%P 381--407
%R doi:10.1023/A:1012978805372
%T Evolving Teams of Predictors with Linear Genetic
Programming
%U http://citeseer.ist.psu.edu/411995.html
%V 2
%X This paper applies the evolution of GP teams to
different classification and regression problems and
compares different methods for combining the outputs of
the team programs. These include hybrid approaches
where (1) a neural network is used to optimize the
weights of programs in a team for a common decision and
(2) a real numbered vector (the representation of
evolution strategies) of weights is evolved with each
term in parallel. The cooperative team approach results
in an improved training and generalization performance
compared to the standard GP method. The higher
computational overhead of team evolution is
counteracted by using a fast variant of linear GP. In
particular, the processing time of linear genetic
programs is reduced significantly by removing intron
code before program execution.
@article{brameier:2001:GPEM,
abstract = {This paper applies the evolution of GP teams to
different classification and regression problems and
compares different methods for combining the outputs of
the team programs. These include hybrid approaches
where (1) a neural network is used to optimize the
weights of programs in a team for a common decision and
(2) a real numbered vector (the representation of
evolution strategies) of weights is evolved with each
term in parallel. The cooperative team approach results
in an improved training and generalization performance
compared to the standard GP method. The higher
computational overhead of team evolution is
counteracted by using a fast variant of linear GP. In
particular, the processing time of linear genetic
programs is reduced significantly by removing intron
code before program execution.},
added-at = {2008-06-19T17:35:00.000+0200},
author = {Brameier, Markus and Banzhaf, Wolfgang},
biburl = {https://www.bibsonomy.org/bibtex/2aa60510b9c8db7f49efd43f45e711655/brazovayeye},
doi = {doi:10.1023/A:1012978805372},
interhash = {620b842f2a69f9e7f6fe3be84d938b40},
intrahash = {aa60510b9c8db7f49efd43f45e711655},
issn = {1389-2576},
journal = {Genetic Programming and Evolvable Machines},
keywords = {algorithms, combination evolution genetic linear multiple of predictors, programming programming, teams,},
month = {December},
notes = {Article ID: 386363},
number = 4,
pages = {381--407},
size = {26 pages},
timestamp = {2008-06-19T17:36:54.000+0200},
title = {Evolving Teams of Predictors with Linear Genetic
Programming},
url = {http://citeseer.ist.psu.edu/411995.html},
volume = 2,
year = 2001
}