Learning monitoring strategies: A difficult genetic programming application

Marc S. Atkin, Paul R Cohen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

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 behaviors. 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.

Original languageEnglish (US)
Title of host publicationIEEE Conference on Evolutionary Computation - Proceedings
PublisherIEEE
Pages328-332
Number of pages5
Volume1
StatePublished - 1994
Externally publishedYes
EventProceedings of the 1st IEEE Conference on Evolutionary Computation. Part 1 (of 2) - Orlando, FL, USA
Duration: Jun 27 1994Jun 29 1994

Other

OtherProceedings of the 1st IEEE Conference on Evolutionary Computation. Part 1 (of 2)
CityOrlando, FL, USA
Period6/27/946/29/94

Fingerprint

Genetic programming
Monitoring
Genetic algorithms

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Atkin, M. S., & Cohen, P. R. (1994). Learning monitoring strategies: A difficult genetic programming application. In IEEE Conference on Evolutionary Computation - Proceedings (Vol. 1, pp. 328-332). IEEE.

Learning monitoring strategies : A difficult genetic programming application. / Atkin, Marc S.; Cohen, Paul R.

IEEE Conference on Evolutionary Computation - Proceedings. Vol. 1 IEEE, 1994. p. 328-332.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Atkin, MS & Cohen, PR 1994, Learning monitoring strategies: A difficult genetic programming application. in IEEE Conference on Evolutionary Computation - Proceedings. vol. 1, IEEE, pp. 328-332, Proceedings of the 1st IEEE Conference on Evolutionary Computation. Part 1 (of 2), Orlando, FL, USA, 6/27/94.
Atkin MS, Cohen PR. Learning monitoring strategies: A difficult genetic programming application. In IEEE Conference on Evolutionary Computation - Proceedings. Vol. 1. IEEE. 1994. p. 328-332
Atkin, Marc S. ; Cohen, Paul R. / Learning monitoring strategies : A difficult genetic programming application. IEEE Conference on Evolutionary Computation - Proceedings. Vol. 1 IEEE, 1994. pp. 328-332
@inproceedings{38757a47b15549c18710394993136102,
title = "Learning monitoring strategies: A difficult genetic programming application",
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 behaviors. 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.",
author = "Atkin, {Marc S.} and Cohen, {Paul R}",
year = "1994",
language = "English (US)",
volume = "1",
pages = "328--332",
booktitle = "IEEE Conference on Evolutionary Computation - Proceedings",
publisher = "IEEE",

}

TY - GEN

T1 - Learning monitoring strategies

T2 - A difficult genetic programming application

AU - Atkin, Marc S.

AU - Cohen, Paul R

PY - 1994

Y1 - 1994

N2 - 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 behaviors. 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.

AB - 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 behaviors. 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.

UR - http://www.scopus.com/inward/record.url?scp=0028594533&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028594533&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0028594533

VL - 1

SP - 328

EP - 332

BT - IEEE Conference on Evolutionary Computation - Proceedings

PB - IEEE

ER -