Removal Bias: a New Cause of Code Growth in Tree Based
Evolutionary Programming
T. Soule, and J. Foster. 1998 IEEE International Conference on Evolutionary
Computation, page 781--186. Anchorage, Alaska, USA, IEEE Press, (5-9 May 1998)
Abstract
This paper presents a new cause of code growth, termed
removal bias. We show that growth due to removal bias
can be expected to occur whenever operations which
remove and replace a variable sized section of code,
e.g. crossover or subtree mutation, are used in an
evolutionary paradigm. Two forms of non-destructive
crossover are used to examine the causes of code
growth. Results support the protective value of
inviable code and removal bias as two distinct causes
of code growth. Both causes of code growth are shown to
exist in at least two different problems.
%0 Conference Paper
%1 soule:1998:rbias
%A Soule, Terence
%A Foster, James A.
%B 1998 IEEE International Conference on Evolutionary
Computation
%C Anchorage, Alaska, USA
%D 1998
%I IEEE Press
%K Code algorithms, bias, genetic growth, length parsimony, programming, removal representations variable
%P 781--186
%T Removal Bias: a New Cause of Code Growth in Tree Based
Evolutionary Programming
%U http://citeseer.ist.psu.edu/313655.html
%X This paper presents a new cause of code growth, termed
removal bias. We show that growth due to removal bias
can be expected to occur whenever operations which
remove and replace a variable sized section of code,
e.g. crossover or subtree mutation, are used in an
evolutionary paradigm. Two forms of non-destructive
crossover are used to examine the causes of code
growth. Results support the protective value of
inviable code and removal bias as two distinct causes
of code growth. Both causes of code growth are shown to
exist in at least two different problems.
@inproceedings{soule:1998:rbias,
abstract = {This paper presents a new cause of code growth, termed
removal bias. We show that growth due to removal bias
can be expected to occur whenever operations which
remove and replace a variable sized section of code,
e.g. crossover or subtree mutation, are used in an
evolutionary paradigm. Two forms of non-destructive
crossover are used to examine the causes of code
growth. Results support the protective value of
inviable code and removal bias as two distinct causes
of code growth. Both causes of code growth are shown to
exist in at least two different problems.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Anchorage, Alaska, USA},
author = {Soule, Terence and Foster, James A.},
biburl = {https://www.bibsonomy.org/bibtex/259a38dd39493834c510d6f688cf71764/brazovayeye},
booktitle = {1998 IEEE International Conference on Evolutionary
Computation},
file = {c136.pdf},
interhash = {b340e72ff1d39a665a822f8d2a4a94e3},
intrahash = {59a38dd39493834c510d6f688cf71764},
keywords = {Code algorithms, bias, genetic growth, length parsimony, programming, removal representations variable},
month = {5-9 May},
notes = {ICEC-98 Held In Conjunction With WCCI-98 --- 1998 IEEE
World Congress on Computational Intelligence},
organisation = {IEEE},
pages = {781--186},
publisher = {IEEE Press},
size = {6 pages},
timestamp = {2008-06-19T17:52:00.000+0200},
title = {Removal Bias: a New Cause of Code Growth in Tree Based
Evolutionary Programming},
url = {http://citeseer.ist.psu.edu/313655.html},
year = 1998
}