Crossover Context in Page-based Linear Genetic
Programming
G. Wilson, and M. Heywood. IEEE CCECE 2002: IEEE Canadian Conference on
Electrical and Computer Engineering, 2, page 809--814. IEEE Press, (12-15 May 2002)
Abstract
This work explores strategy learning through genetic
programming in artificial ants that navigate the San
Mateo trail. We investigate several properties of
linearly structured (as opposed to typical tree based)
GP including: the significance of simple register based
memories, the significance of constraints applied to
the crossover operator, and how active the ant are. We
also provide a basis for investigating more thoroughly
the relation between specific code sequences and
fitness by dividing the genome into pages of
instructions and introducing an analysis of fitness
change and exploration of the trail done by particular
parts of a genome. By doing so we are able to present
results on how best to find the instructions in an
individual's program that contribute positively to the
accumulation of effective search strategies.
%0 Conference Paper
%1 wilson:2003:ccpb
%A Wilson, G. C.
%A Heywood, M. I.
%B IEEE CCECE 2002: IEEE Canadian Conference on
Electrical and Computer Engineering
%D 2002
%E Kinsner, W.
%E Seback, A.
%E Ferens, K.
%I IEEE Press
%K (artificial Learning, Mateo San Strategy algorithms, ants, artificial based change, code crossover effective fitness genetic instructions, intelligence), learning memories, operator, problems, programming, register search sequences, simple strategies, strategy trail,
%P 809--814
%T Crossover Context in Page-based Linear Genetic
Programming
%U http://flame.cs.dal.ca/~gwilson/docs/papers/ccece_2002.pdf
%V 2
%X This work explores strategy learning through genetic
programming in artificial ants that navigate the San
Mateo trail. We investigate several properties of
linearly structured (as opposed to typical tree based)
GP including: the significance of simple register based
memories, the significance of constraints applied to
the crossover operator, and how active the ant are. We
also provide a basis for investigating more thoroughly
the relation between specific code sequences and
fitness by dividing the genome into pages of
instructions and introducing an analysis of fitness
change and exploration of the trail done by particular
parts of a genome. By doing so we are able to present
results on how best to find the instructions in an
individual's program that contribute positively to the
accumulation of effective search strategies.
%@ 0-7803-7515-7
@inproceedings{wilson:2003:ccpb,
abstract = {This work explores strategy learning through genetic
programming in artificial ants that navigate the San
Mateo trail. We investigate several properties of
linearly structured (as opposed to typical tree based)
GP including: the significance of simple register based
memories, the significance of constraints applied to
the crossover operator, and how active the ant are. We
also provide a basis for investigating more thoroughly
the relation between specific code sequences and
fitness by dividing the genome into pages of
instructions and introducing an analysis of fitness
change and exploration of the trail done by particular
parts of a genome. By doing so we are able to present
results on how best to find the instructions in an
individual's program that contribute positively to the
accumulation of effective search strategies.},
added-at = {2008-06-19T17:35:00.000+0200},
author = {Wilson, G. C. and Heywood, M. I.},
biburl = {https://www.bibsonomy.org/bibtex/22c72016f4f2ca6fc4b01d5cc1ef4873e/brazovayeye},
booktitle = {IEEE CCECE 2002: IEEE Canadian Conference on
Electrical and Computer Engineering},
editor = {Kinsner, W. and Seback, A. and Ferens, K.},
interhash = {3198643db5b89a35cca1e38cb3718c10},
intrahash = {2c72016f4f2ca6fc4b01d5cc1ef4873e},
isbn = {0-7803-7515-7},
issn = {0840-7789},
keywords = {(artificial Learning, Mateo San Strategy algorithms, ants, artificial based change, code crossover effective fitness genetic instructions, intelligence), learning memories, operator, problems, programming, register search sequences, simple strategies, strategy trail,},
month = {12-15 May},
notes = {best student paper award},
organisation = {IEEE Canada},
pages = {809--814},
publisher = {IEEE Press},
size = {6 pages},
timestamp = {2008-06-19T17:54:13.000+0200},
title = {Crossover Context in Page-based Linear Genetic
Programming},
url = {http://flame.cs.dal.ca/~gwilson/docs/papers/ccece_2002.pdf},
volume = 2,
year = 2002
}