Speeding up evolution through learning

LEM

Ryszard S. Michalski, Guido Cervone, Kenneth Kaufman

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

6 Citations (Scopus)

Abstract

This paper reports briefly on the development of a new approach to evolutionary computation, called the Learnable Evolution Model or LEM. In contrast to conventional Darwinian-type evolutionary algorithms that employ mutation and/or recombination, LEM employs machine learning to generate new populations. At each step of evolution, LEM determines hypotheses explaining why certain individuals in the population are superior to others in performing the designated class of tasks. These hypotheses are then instantiated to create a next generation. In the testing studies described here, we compared a program implementing LEM with selected evolutionary computation algorithms on a range optimization problems and a filter design problem. In these studies, LEM significantly outperformed the evolutionary computation algorithms, sometimes speeding up the evolution by two or more orders of magnitude in the number of evolutionary steps (births). LEM was also applied to a real-world problem of designing optimized heat exchangers. The resulting designs matched or - outperformed the best human designs.

Original languageEnglish (US)
Title of host publicationIntelligent Information Systems -Proceedings of the IIS'2000 Symposium
Pages243-256
Number of pages14
EditionAISC
DOIs
StatePublished - Dec 1 2000
Event9th Intelligent Information Systems Symposium, IIS'2000 - Bystra, Poland
Duration: Jun 12 2000Jun 16 2000

Publication series

NameAdvances in Soft Computing
NumberAISC
Volume4
ISSN (Print)1615-3871
ISSN (Electronic)1860-0794

Other

Other9th Intelligent Information Systems Symposium, IIS'2000
CountryPoland
CityBystra
Period6/12/006/16/00

Fingerprint

Evolutionary algorithms
Heat exchangers
Learning systems
Testing

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)
  • Computational Mechanics
  • Computer Science Applications

Cite this

Michalski, R. S., Cervone, G., & Kaufman, K. (2000). Speeding up evolution through learning: LEM. In Intelligent Information Systems -Proceedings of the IIS'2000 Symposium (AISC ed., pp. 243-256). (Advances in Soft Computing; Vol. 4, No. AISC). https://doi.org/10.1007/978-3-7908-1846-8_22
Michalski, Ryszard S. ; Cervone, Guido ; Kaufman, Kenneth. / Speeding up evolution through learning : LEM. Intelligent Information Systems -Proceedings of the IIS'2000 Symposium. AISC. ed. 2000. pp. 243-256 (Advances in Soft Computing; AISC).
@inproceedings{ccf44edc38ca40569e99337c7d27401f,
title = "Speeding up evolution through learning: LEM",
abstract = "This paper reports briefly on the development of a new approach to evolutionary computation, called the Learnable Evolution Model or LEM. In contrast to conventional Darwinian-type evolutionary algorithms that employ mutation and/or recombination, LEM employs machine learning to generate new populations. At each step of evolution, LEM determines hypotheses explaining why certain individuals in the population are superior to others in performing the designated class of tasks. These hypotheses are then instantiated to create a next generation. In the testing studies described here, we compared a program implementing LEM with selected evolutionary computation algorithms on a range optimization problems and a filter design problem. In these studies, LEM significantly outperformed the evolutionary computation algorithms, sometimes speeding up the evolution by two or more orders of magnitude in the number of evolutionary steps (births). LEM was also applied to a real-world problem of designing optimized heat exchangers. The resulting designs matched or - outperformed the best human designs.",
author = "Michalski, {Ryszard S.} and Guido Cervone and Kenneth Kaufman",
year = "2000",
month = "12",
day = "1",
doi = "10.1007/978-3-7908-1846-8_22",
language = "English (US)",
isbn = "9783790813098",
series = "Advances in Soft Computing",
number = "AISC",
pages = "243--256",
booktitle = "Intelligent Information Systems -Proceedings of the IIS'2000 Symposium",
edition = "AISC",

}

Michalski, RS, Cervone, G & Kaufman, K 2000, Speeding up evolution through learning: LEM. in Intelligent Information Systems -Proceedings of the IIS'2000 Symposium. AISC edn, Advances in Soft Computing, no. AISC, vol. 4, pp. 243-256, 9th Intelligent Information Systems Symposium, IIS'2000, Bystra, Poland, 6/12/00. https://doi.org/10.1007/978-3-7908-1846-8_22

Speeding up evolution through learning : LEM. / Michalski, Ryszard S.; Cervone, Guido; Kaufman, Kenneth.

Intelligent Information Systems -Proceedings of the IIS'2000 Symposium. AISC. ed. 2000. p. 243-256 (Advances in Soft Computing; Vol. 4, No. AISC).

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

TY - GEN

T1 - Speeding up evolution through learning

T2 - LEM

AU - Michalski, Ryszard S.

AU - Cervone, Guido

AU - Kaufman, Kenneth

PY - 2000/12/1

Y1 - 2000/12/1

N2 - This paper reports briefly on the development of a new approach to evolutionary computation, called the Learnable Evolution Model or LEM. In contrast to conventional Darwinian-type evolutionary algorithms that employ mutation and/or recombination, LEM employs machine learning to generate new populations. At each step of evolution, LEM determines hypotheses explaining why certain individuals in the population are superior to others in performing the designated class of tasks. These hypotheses are then instantiated to create a next generation. In the testing studies described here, we compared a program implementing LEM with selected evolutionary computation algorithms on a range optimization problems and a filter design problem. In these studies, LEM significantly outperformed the evolutionary computation algorithms, sometimes speeding up the evolution by two or more orders of magnitude in the number of evolutionary steps (births). LEM was also applied to a real-world problem of designing optimized heat exchangers. The resulting designs matched or - outperformed the best human designs.

AB - This paper reports briefly on the development of a new approach to evolutionary computation, called the Learnable Evolution Model or LEM. In contrast to conventional Darwinian-type evolutionary algorithms that employ mutation and/or recombination, LEM employs machine learning to generate new populations. At each step of evolution, LEM determines hypotheses explaining why certain individuals in the population are superior to others in performing the designated class of tasks. These hypotheses are then instantiated to create a next generation. In the testing studies described here, we compared a program implementing LEM with selected evolutionary computation algorithms on a range optimization problems and a filter design problem. In these studies, LEM significantly outperformed the evolutionary computation algorithms, sometimes speeding up the evolution by two or more orders of magnitude in the number of evolutionary steps (births). LEM was also applied to a real-world problem of designing optimized heat exchangers. The resulting designs matched or - outperformed the best human designs.

UR - http://www.scopus.com/inward/record.url?scp=24344442702&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=24344442702&partnerID=8YFLogxK

U2 - 10.1007/978-3-7908-1846-8_22

DO - 10.1007/978-3-7908-1846-8_22

M3 - Conference contribution

SN - 9783790813098

T3 - Advances in Soft Computing

SP - 243

EP - 256

BT - Intelligent Information Systems -Proceedings of the IIS'2000 Symposium

ER -

Michalski RS, Cervone G, Kaufman K. Speeding up evolution through learning: LEM. In Intelligent Information Systems -Proceedings of the IIS'2000 Symposium. AISC ed. 2000. p. 243-256. (Advances in Soft Computing; AISC). https://doi.org/10.1007/978-3-7908-1846-8_22