Prospects for Computational Steering of Evolutionary Computation

Seth Bullock; John P Cartlidge; M. Thompson

Prospects for Computational Steering of Evolutionary Computation

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

196 Downloads (Pure)

Abstract

Currently, evolutionary computation (EC) typically takes place in batch mode: algorithms are run autonomously, with the user providing little or no intervention or guidance. Although it is rarely possible to specify in advance, on the basis of EC theory, the optimal evolutionary algorithm for a particular problem, it seems likely that experienced EC practitioners possess considerable tacit knowledge of how evolutionary algorithms work. In situations such as this, computational steering (ongoing, informed user intervention in the execution of an otherwise autonomous computational process) has been profitably exploited to improve performance and generate insights into computational processes. In this short paper, prospects for the computational steering of evolutionary computation are assessed, and a prototype example of computational steering applied to a coevolutionary algorithm is presented.

Original language	English
Title of host publication	Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life
Subtitle of host publication	Beyond Fitness - Visualising Evolution
Editors	E. Bilotta, Gro D., T. Smith, T. Lenaerts, S. Bullock, H. H. Lund, J. Bird, R. Watson, P. Pantano, L. Pagliarini, H. Abbass, R. Standish, M. Bedau
Place of Publication	Sydney, Australia
Publisher	Massachusetts Institute of Technology (MIT) Press
Pages	131-137
Number of pages	7
Publication status	Published - Dec 2002
Event	8th International Conference on Artificial Life (ALife VIII) - Sydney, Australia Duration: 9 Dec 2002 → 13 Dec 2002

Conference

Conference	8th International Conference on Artificial Life (ALife VIII)
Country/Territory	Australia
City	Sydney
Period	9/12/02 → 13/12/02

Keywords

evolutionary computation
coevolution
evolutionary steering
virulence
visualisation
reduced virulence
virulence adaptation
disengagement
coevolutionary disengagement
coevolutionary dynamics

Access to Document

Prospects for Computational Steering of Evolutionary ComputationAccepted author manuscript, 157 KB

Handle.net

Persistent link

Cite this

Bullock, S., Cartlidge, J. P., & Thompson, M. (2002). Prospects for Computational Steering of Evolutionary Computation. In E. Bilotta, G. D., T. Smith, T. Lenaerts, S. Bullock, H. H. Lund, J. Bird, R. Watson, P. Pantano, L. Pagliarini, H. Abbass, R. Standish, & M. Bedau (Eds.), Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life: Beyond Fitness - Visualising Evolution (pp. 131-137). Massachusetts Institute of Technology (MIT) Press.

Bullock, Seth ; Cartlidge, John P ; Thompson, M. / Prospects for Computational Steering of Evolutionary Computation. Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life: Beyond Fitness - Visualising Evolution. editor / E. Bilotta ; Gro D. ; T. Smith ; T. Lenaerts ; S. Bullock ; H. H. Lund ; J. Bird ; R. Watson ; P. Pantano ; L. Pagliarini ; H. Abbass ; R. Standish ; M. Bedau. Sydney, Australia : Massachusetts Institute of Technology (MIT) Press, 2002. pp. 131-137

@inproceedings{7fe7a8930c064c32affdfc44cf243532,

title = "Prospects for Computational Steering of Evolutionary Computation",

abstract = "Currently, evolutionary computation (EC) typically takes place in batch mode: algorithms are run autonomously, with the user providing little or no intervention or guidance. Although it is rarely possible to specify in advance, on the basis of EC theory, the optimal evolutionary algorithm for a particular problem, it seems likely that experienced EC practitioners possess considerable tacit knowledge of how evolutionary algorithms work. In situations such as this, computational steering (ongoing, informed user intervention in the execution of an otherwise autonomous computational process) has been profitably exploited to improve performance and generate insights into computational processes. In this short paper, prospects for the computational steering of evolutionary computation are assessed, and a prototype example of computational steering applied to a coevolutionary algorithm is presented.",

keywords = "evolutionary computation, coevolution, evolutionary steering, virulence, visualisation, reduced virulence, virulence adaptation, disengagement, coevolutionary disengagement, coevolutionary dynamics",

author = "Seth Bullock and Cartlidge, \{John P\} and M. Thompson",

year = "2002",

month = dec,

language = "English",

pages = "131--137",

editor = "E. Bilotta and Gro D. and T. Smith and T. Lenaerts and S. Bullock and Lund, \{H. H.\} and J. Bird and R. Watson and P. Pantano and L. Pagliarini and H. Abbass and R. Standish and M. Bedau",

booktitle = "Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life",

publisher = "Massachusetts Institute of Technology (MIT) Press",

address = "United States",

note = "8th International Conference on Artificial Life (ALife VIII) ; Conference date: 09-12-2002 Through 13-12-2002",

}

Bullock, S , Cartlidge, JP & Thompson, M 2002, Prospects for Computational Steering of Evolutionary Computation. in E Bilotta, G D., T Smith, T Lenaerts, S Bullock, HH Lund, J Bird, R Watson, P Pantano, L Pagliarini, H Abbass, R Standish & M Bedau (eds), Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life: Beyond Fitness - Visualising Evolution. Massachusetts Institute of Technology (MIT) Press, Sydney, Australia, pp. 131-137, 8th International Conference on Artificial Life (ALife VIII), Sydney, Australia, 9/12/02.

Prospects for Computational Steering of Evolutionary Computation. / Bullock, Seth ; Cartlidge, John P; Thompson, M.
Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life: Beyond Fitness - Visualising Evolution. ed. / E. Bilotta; Gro D.; T. Smith; T. Lenaerts; S. Bullock; H. H. Lund; J. Bird; R. Watson; P. Pantano; L. Pagliarini; H. Abbass; R. Standish; M. Bedau. Sydney, Australia: Massachusetts Institute of Technology (MIT) Press, 2002. p. 131-137.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Prospects for Computational Steering of Evolutionary Computation

AU - Bullock, Seth

AU - Cartlidge, John P

AU - Thompson, M.

PY - 2002/12

Y1 - 2002/12

N2 - Currently, evolutionary computation (EC) typically takes place in batch mode: algorithms are run autonomously, with the user providing little or no intervention or guidance. Although it is rarely possible to specify in advance, on the basis of EC theory, the optimal evolutionary algorithm for a particular problem, it seems likely that experienced EC practitioners possess considerable tacit knowledge of how evolutionary algorithms work. In situations such as this, computational steering (ongoing, informed user intervention in the execution of an otherwise autonomous computational process) has been profitably exploited to improve performance and generate insights into computational processes. In this short paper, prospects for the computational steering of evolutionary computation are assessed, and a prototype example of computational steering applied to a coevolutionary algorithm is presented.

AB - Currently, evolutionary computation (EC) typically takes place in batch mode: algorithms are run autonomously, with the user providing little or no intervention or guidance. Although it is rarely possible to specify in advance, on the basis of EC theory, the optimal evolutionary algorithm for a particular problem, it seems likely that experienced EC practitioners possess considerable tacit knowledge of how evolutionary algorithms work. In situations such as this, computational steering (ongoing, informed user intervention in the execution of an otherwise autonomous computational process) has been profitably exploited to improve performance and generate insights into computational processes. In this short paper, prospects for the computational steering of evolutionary computation are assessed, and a prototype example of computational steering applied to a coevolutionary algorithm is presented.

KW - evolutionary computation

KW - coevolution

KW - evolutionary steering

KW - virulence

KW - visualisation

KW - reduced virulence

KW - virulence adaptation

KW - disengagement

KW - coevolutionary disengagement

KW - coevolutionary dynamics

UR - http://seis.bris.ac.uk/~sb15704/pubs_abstracts.html#eps261459

M3 - Conference Contribution (Conference Proceeding)

SP - 131

EP - 137

BT - Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life

A2 - Bilotta, E.

A2 - D., Gro

A2 - Smith, T.

A2 - Lenaerts, T.

A2 - Bullock, S.

A2 - Lund, H. H.

A2 - Bird, J.

A2 - Watson, R.

A2 - Pantano, P.

A2 - Pagliarini, L.

A2 - Abbass, H.

A2 - Standish, R.

A2 - Bedau, M.

PB - Massachusetts Institute of Technology (MIT) Press

CY - Sydney, Australia

T2 - 8th International Conference on Artificial Life (ALife VIII)

Y2 - 9 December 2002 through 13 December 2002

ER -

Bullock S , Cartlidge JP, Thompson M. Prospects for Computational Steering of Evolutionary Computation. In Bilotta E, D. G, Smith T, Lenaerts T, Bullock S, Lund HH, Bird J, Watson R, Pantano P, Pagliarini L, Abbass H, Standish R, Bedau M, editors, Artificial Life VIII: Workshop Proceedings of the Eighth International Conference on Artificial Life: Beyond Fitness - Visualising Evolution. Sydney, Australia: Massachusetts Institute of Technology (MIT) Press. 2002. p. 131-137

Prospects for Computational Steering of Evolutionary Computation

Abstract

Conference

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this