Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation

Eduardo Segredo, Eduardo Lalla-Ruiz, Emma Hart, Ben Paechter, Stefan Voß

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    4 Citations (Scopus)

    Abstract

    Choosing the correct algorithm to solve a problem still remains an issue 40 years after the Algorithm Selection Problem was first posed. Here we propose a hyper-heuristic which can apply one of two meta-heuristics at the current stage of the search. A scoring function is used to select the most appropriate algorithm based on an estimate of the improvement that might be made by applying each algorithm. We use a differential evolution algorithm and a genetic algorithm as the two metaheuristics and assess performance on a suite of 18 functions provided by the Generalization-based Contest in Global Optimization (genopt). The experimental evaluation shows that the hybridisation is able to provide an improvement with respect to the results obtained by both the differential evolution scheme and the genetic algorithm when they are executed independently. In addition, the high performance of our hybrid approach allowed two out of the three prizes available at genopt to be obtained.

    Original languageEnglish
    Title of host publicationLearning and Intelligent Optimization
    Subtitle of host publication10th International Conference, LION 10, Ischia, Italy, May 29 - June 1, 2016, Revised Selected Papers
    EditorsPaola Festa, Meinolf Sellmann, Joaquin Vanschoren
    PublisherSpringer
    Pages296-305
    Number of pages10
    ISBN (Electronic)978-3-319-50349-3
    ISBN (Print)978-3-319-50348-6
    DOIs
    Publication statusPublished - 1 Jan 2016
    Event10th International Conference on Learning and Intelligent Optimization, LION 2016 - Ischia, Italy
    Duration: 29 May 20161 Jun 2016
    Conference number: 10

    Publication series

    NameLecture Notes in Computer Science
    Volume10079
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349
    NameLecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics
    PublisherSpringer

    Conference

    Conference10th International Conference on Learning and Intelligent Optimization, LION 2016
    Abbreviated titleLION
    CountryItaly
    CityIschia
    Period29/05/161/06/16

    Fingerprint

    Hyper-heuristics
    Continuous Optimization
    Evolutionary algorithms
    Global Optimization
    Evolutionary Algorithms
    Metaheuristics
    Genetic Algorithm
    Genetic algorithms
    Differential Evolution Algorithm
    Hybrid Approach
    Differential Evolution
    Scoring
    Experimental Evaluation
    Global optimization
    High Performance
    Hybridization
    Heuristics
    Estimate

    Keywords

    • Differential evolution
    • Genetic algorithm
    • Global continuous optimization
    • Global search
    • Hyper-heuristic

    Cite this

    Segredo, E., Lalla-Ruiz, E., Hart, E., Paechter, B., & Voß, S. (2016). Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation. In P. Festa, M. Sellmann, & J. Vanschoren (Eds.), Learning and Intelligent Optimization: 10th International Conference, LION 10, Ischia, Italy, May 29 - June 1, 2016, Revised Selected Papers (pp. 296-305). (Lecture Notes in Computer Science; Vol. 10079), (Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Springer. https://doi.org/10.1007/978-3-319-50349-3_25
    Segredo, Eduardo ; Lalla-Ruiz, Eduardo ; Hart, Emma ; Paechter, Ben ; Voß, Stefan. / Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation. Learning and Intelligent Optimization: 10th International Conference, LION 10, Ischia, Italy, May 29 - June 1, 2016, Revised Selected Papers. editor / Paola Festa ; Meinolf Sellmann ; Joaquin Vanschoren. Springer, 2016. pp. 296-305 (Lecture Notes in Computer Science). (Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).
    @inproceedings{c20d7708ea974ee09c74a354a03bf077,
    title = "Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation",
    abstract = "Choosing the correct algorithm to solve a problem still remains an issue 40 years after the Algorithm Selection Problem was first posed. Here we propose a hyper-heuristic which can apply one of two meta-heuristics at the current stage of the search. A scoring function is used to select the most appropriate algorithm based on an estimate of the improvement that might be made by applying each algorithm. We use a differential evolution algorithm and a genetic algorithm as the two metaheuristics and assess performance on a suite of 18 functions provided by the Generalization-based Contest in Global Optimization (genopt). The experimental evaluation shows that the hybridisation is able to provide an improvement with respect to the results obtained by both the differential evolution scheme and the genetic algorithm when they are executed independently. In addition, the high performance of our hybrid approach allowed two out of the three prizes available at genopt to be obtained.",
    keywords = "Differential evolution, Genetic algorithm, Global continuous optimization, Global search, Hyper-heuristic",
    author = "Eduardo Segredo and Eduardo Lalla-Ruiz and Emma Hart and Ben Paechter and Stefan Vo{\ss}",
    year = "2016",
    month = "1",
    day = "1",
    doi = "10.1007/978-3-319-50349-3_25",
    language = "English",
    isbn = "978-3-319-50348-6",
    series = "Lecture Notes in Computer Science",
    publisher = "Springer",
    pages = "296--305",
    editor = "Festa, {Paola } and Meinolf Sellmann and Joaquin Vanschoren",
    booktitle = "Learning and Intelligent Optimization",

    }

    Segredo, E, Lalla-Ruiz, E, Hart, E, Paechter, B & Voß, S 2016, Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation. in P Festa, M Sellmann & J Vanschoren (eds), Learning and Intelligent Optimization: 10th International Conference, LION 10, Ischia, Italy, May 29 - June 1, 2016, Revised Selected Papers. Lecture Notes in Computer Science, vol. 10079, Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics, Springer, pp. 296-305, 10th International Conference on Learning and Intelligent Optimization, LION 2016, Ischia, Italy, 29/05/16. https://doi.org/10.1007/978-3-319-50349-3_25

    Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation. / Segredo, Eduardo; Lalla-Ruiz, Eduardo; Hart, Emma; Paechter, Ben; Voß, Stefan.

    Learning and Intelligent Optimization: 10th International Conference, LION 10, Ischia, Italy, May 29 - June 1, 2016, Revised Selected Papers. ed. / Paola Festa; Meinolf Sellmann; Joaquin Vanschoren. Springer, 2016. p. 296-305 (Lecture Notes in Computer Science; Vol. 10079), (Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    TY - GEN

    T1 - Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation

    AU - Segredo, Eduardo

    AU - Lalla-Ruiz, Eduardo

    AU - Hart, Emma

    AU - Paechter, Ben

    AU - Voß, Stefan

    PY - 2016/1/1

    Y1 - 2016/1/1

    N2 - Choosing the correct algorithm to solve a problem still remains an issue 40 years after the Algorithm Selection Problem was first posed. Here we propose a hyper-heuristic which can apply one of two meta-heuristics at the current stage of the search. A scoring function is used to select the most appropriate algorithm based on an estimate of the improvement that might be made by applying each algorithm. We use a differential evolution algorithm and a genetic algorithm as the two metaheuristics and assess performance on a suite of 18 functions provided by the Generalization-based Contest in Global Optimization (genopt). The experimental evaluation shows that the hybridisation is able to provide an improvement with respect to the results obtained by both the differential evolution scheme and the genetic algorithm when they are executed independently. In addition, the high performance of our hybrid approach allowed two out of the three prizes available at genopt to be obtained.

    AB - Choosing the correct algorithm to solve a problem still remains an issue 40 years after the Algorithm Selection Problem was first posed. Here we propose a hyper-heuristic which can apply one of two meta-heuristics at the current stage of the search. A scoring function is used to select the most appropriate algorithm based on an estimate of the improvement that might be made by applying each algorithm. We use a differential evolution algorithm and a genetic algorithm as the two metaheuristics and assess performance on a suite of 18 functions provided by the Generalization-based Contest in Global Optimization (genopt). The experimental evaluation shows that the hybridisation is able to provide an improvement with respect to the results obtained by both the differential evolution scheme and the genetic algorithm when they are executed independently. In addition, the high performance of our hybrid approach allowed two out of the three prizes available at genopt to be obtained.

    KW - Differential evolution

    KW - Genetic algorithm

    KW - Global continuous optimization

    KW - Global search

    KW - Hyper-heuristic

    UR - http://www.scopus.com/inward/record.url?scp=85006839828&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-50349-3_25

    DO - 10.1007/978-3-319-50349-3_25

    M3 - Conference contribution

    SN - 978-3-319-50348-6

    T3 - Lecture Notes in Computer Science

    SP - 296

    EP - 305

    BT - Learning and Intelligent Optimization

    A2 - Festa, Paola

    A2 - Sellmann, Meinolf

    A2 - Vanschoren, Joaquin

    PB - Springer

    ER -

    Segredo E, Lalla-Ruiz E, Hart E, Paechter B, Voß S. Hybridisation of evolutionary algorithms through hyper-heuristics for global continuous optimisation. In Festa P, Sellmann M, Vanschoren J, editors, Learning and Intelligent Optimization: 10th International Conference, LION 10, Ischia, Italy, May 29 - June 1, 2016, Revised Selected Papers. Springer. 2016. p. 296-305. (Lecture Notes in Computer Science). (Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). https://doi.org/10.1007/978-3-319-50349-3_25