Learning monitoring strategies: A difficult genetic
programming application
M. Atkin, and P. Cohen. Proceedings of the 1994 IEEE World Congress on
Computational Intelligence, page 328--332a. Orlando, Florida, USA, IEEE Press, (27-29 June 1994)
Abstract
Finding optimal or at least good monitoring strategies
is an important consideration when designing an agent.
We have applied genetic programming to this task, with
mixed results. Since the agent control language was
kept purposefully general, the set of monitoring
strategies constitutes only a small part of the overall
space of possible behaviours. Because of this, it was
often difficult for the genetic algorithm to evolve
them, even though their performance was superior. These
results raise questions as to how easy it will be for
genetic programming to scale up as the areas it is
applied to become more complex.
Proceedings of the 1994 IEEE World Congress on
Computational Intelligence
year
1994
month
27-29 June
pages
328--332a
publisher
IEEE Press
size
6 pages
notes
Novel? chrome/program structure linear, close to
assembly lanuage, used GOTOs and interrupt handlers.
Did _not_ get performance improvement on changing to
parse trees. Did evolve progs to control agents which
moved to the goal without colliding with an obstacle.
Finally cautions about problems with GP scaling
up.
Älso tried local mating (also known as fine grain
parallelism)"
Also available as Technical Report 94-52, Dept. of
Computer Science, University of Massachusetts/Amherst,
USA?
%0 Conference Paper
%1 Atkin:1994:LMSDGP
%A Atkin, Marc S.
%A Cohen, Paul R.
%B Proceedings of the 1994 IEEE World Congress on
Computational Intelligence
%C Orlando, Florida, USA
%D 1994
%I IEEE Press
%K algorithms, cupcake genetic problem programming,
%P 328--332a
%T Learning monitoring strategies: A difficult genetic
programming application
%U http://citeseer.ist.psu.edu/94049.html
%X Finding optimal or at least good monitoring strategies
is an important consideration when designing an agent.
We have applied genetic programming to this task, with
mixed results. Since the agent control language was
kept purposefully general, the set of monitoring
strategies constitutes only a small part of the overall
space of possible behaviours. Because of this, it was
often difficult for the genetic algorithm to evolve
them, even though their performance was superior. These
results raise questions as to how easy it will be for
genetic programming to scale up as the areas it is
applied to become more complex.
@inproceedings{Atkin:1994:LMSDGP,
abstract = {Finding optimal or at least good monitoring strategies
is an important consideration when designing an agent.
We have applied genetic programming to this task, with
mixed results. Since the agent control language was
kept purposefully general, the set of monitoring
strategies constitutes only a small part of the overall
space of possible behaviours. Because of this, it was
often difficult for the genetic algorithm to evolve
them, even though their performance was superior. These
results raise questions as to how easy it will be for
genetic programming to scale up as the areas it is
applied to become more complex.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Orlando, Florida, USA},
author = {Atkin, Marc S. and Cohen, Paul R.},
biburl = {https://www.bibsonomy.org/bibtex/2b76fdec65cc7b7eaee7e802755286551/brazovayeye},
booktitle = {Proceedings of the 1994 IEEE World Congress on
Computational Intelligence},
interhash = {dbf79703a367f76a5d03dcb2916dc546},
intrahash = {b76fdec65cc7b7eaee7e802755286551},
keywords = {algorithms, cupcake genetic problem programming,},
month = {27-29 June},
notes = {Novel? chrome/program structure linear, close to
assembly lanuage, used GOTOs and interrupt handlers.
Did _not_ get performance improvement on changing to
parse trees. Did evolve progs to control agents which
moved to the goal without colliding with an obstacle.
Finally cautions about problems with GP scaling
up.
{"}Also tried local mating (also known as fine grain
parallelism){"}
Also available as Technical Report 94-52, Dept. of
Computer Science, University of Massachusetts/Amherst,
USA?},
pages = {328--332a},
publisher = {IEEE Press},
size = {6 pages},
timestamp = {2008-06-19T17:35:53.000+0200},
title = {Learning monitoring strategies: {A} difficult genetic
programming application},
url = {http://citeseer.ist.psu.edu/94049.html},
year = 1994
}