An evaluation of off-line calibration techniques for evolutionary algorithms

Elizabeth Montero, María Cristina Riff, Bertrand Neveu

Resultado de la investigación: Conference contribution

5 Citas (Scopus)

Resumen

Most metaheuristics define a set of parameters that must be tuned. A good setup of those parameter values can lead to take advantage of all the metaheuristic capabilities to solve the problem at hand. Tuning techniques are step by step methods based on multiple runs of the algorithm. In this study we compare three automated tuning methods: FRace, Revac and ParamILS. We evaluate the performance of each method using a genetic algorithm for combinatorial optimization. The differences and advantages of each technique are discussed. Finally we establish some guidelines that might help to choose a tuning process to use.

Idioma originalEnglish
Título de la publicación alojadaProceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10
Páginas299-300
Número de páginas2
DOI
EstadoPublished - 27 ago 2010
Evento12th Annual Genetic and Evolutionary Computation Conference, GECCO-2010 - Portland, OR, United States
Duración: 7 jul 201011 jul 2010

Conference

Conference12th Annual Genetic and Evolutionary Computation Conference, GECCO-2010
PaísUnited States
CiudadPortland, OR
Período7/07/1011/07/10

Huella dactilar

Evolutionary algorithms
Evolutionary Algorithms
Tuning
Calibration
Metaheuristics
Line
Evaluation
Combinatorial optimization
Combinatorial Optimization
Choose
Genetic algorithms
Genetic Algorithm
Evaluate

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Theoretical Computer Science

Citar esto

Montero, E., Riff, M. C., & Neveu, B. (2010). An evaluation of off-line calibration techniques for evolutionary algorithms. En Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10 (pp. 299-300) https://doi.org/10.1145/1830483.1830540
Montero, Elizabeth ; Riff, María Cristina ; Neveu, Bertrand. / An evaluation of off-line calibration techniques for evolutionary algorithms. Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. 2010. pp. 299-300
@inproceedings{20812c188c5d470a820524ccfc48575d,
title = "An evaluation of off-line calibration techniques for evolutionary algorithms",
abstract = "Most metaheuristics define a set of parameters that must be tuned. A good setup of those parameter values can lead to take advantage of all the metaheuristic capabilities to solve the problem at hand. Tuning techniques are step by step methods based on multiple runs of the algorithm. In this study we compare three automated tuning methods: FRace, Revac and ParamILS. We evaluate the performance of each method using a genetic algorithm for combinatorial optimization. The differences and advantages of each technique are discussed. Finally we establish some guidelines that might help to choose a tuning process to use.",
keywords = "Evolutionary algorithms, Parameter setting problem",
author = "Elizabeth Montero and Riff, {Mar{\'i}a Cristina} and Bertrand Neveu",
year = "2010",
month = "8",
day = "27",
doi = "10.1145/1830483.1830540",
language = "English",
isbn = "9781450300728",
pages = "299--300",
booktitle = "Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10",

}

Montero, E, Riff, MC & Neveu, B 2010, An evaluation of off-line calibration techniques for evolutionary algorithms. En Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. pp. 299-300, 12th Annual Genetic and Evolutionary Computation Conference, GECCO-2010, Portland, OR, United States, 7/07/10. https://doi.org/10.1145/1830483.1830540

An evaluation of off-line calibration techniques for evolutionary algorithms. / Montero, Elizabeth; Riff, María Cristina; Neveu, Bertrand.

Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. 2010. p. 299-300.

Resultado de la investigación: Conference contribution

TY - GEN

T1 - An evaluation of off-line calibration techniques for evolutionary algorithms

AU - Montero, Elizabeth

AU - Riff, María Cristina

AU - Neveu, Bertrand

PY - 2010/8/27

Y1 - 2010/8/27

N2 - Most metaheuristics define a set of parameters that must be tuned. A good setup of those parameter values can lead to take advantage of all the metaheuristic capabilities to solve the problem at hand. Tuning techniques are step by step methods based on multiple runs of the algorithm. In this study we compare three automated tuning methods: FRace, Revac and ParamILS. We evaluate the performance of each method using a genetic algorithm for combinatorial optimization. The differences and advantages of each technique are discussed. Finally we establish some guidelines that might help to choose a tuning process to use.

AB - Most metaheuristics define a set of parameters that must be tuned. A good setup of those parameter values can lead to take advantage of all the metaheuristic capabilities to solve the problem at hand. Tuning techniques are step by step methods based on multiple runs of the algorithm. In this study we compare three automated tuning methods: FRace, Revac and ParamILS. We evaluate the performance of each method using a genetic algorithm for combinatorial optimization. The differences and advantages of each technique are discussed. Finally we establish some guidelines that might help to choose a tuning process to use.

KW - Evolutionary algorithms

KW - Parameter setting problem

UR - http://www.scopus.com/inward/record.url?scp=77955901415&partnerID=8YFLogxK

U2 - 10.1145/1830483.1830540

DO - 10.1145/1830483.1830540

M3 - Conference contribution

SN - 9781450300728

SP - 299

EP - 300

BT - Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10

ER -

Montero E, Riff MC, Neveu B. An evaluation of off-line calibration techniques for evolutionary algorithms. En Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. 2010. p. 299-300 https://doi.org/10.1145/1830483.1830540