Features for Exploiting Black-Box Optimization Problem Structure.

Kevin Tierney, Yuri Malitsky, Tinus Abell

    Research output: Conference Article in Proceeding or Book/Report chapterBook chapterResearchpeer-review


    Black-box optimization (BBO) problems arise in numerous
    scientic and engineering applications and are characterized by compu-
    tationally intensive objective functions, which severely limit the number
    of evaluations that can be performed. We present a robust set of features
    that analyze the tness landscape of BBO problems and show how an
    algorithm portfolio approach can exploit these general, problem indepen-
    dent features and outperform the utilization of any single minimization
    search strategy. We test our methodology on data from the GECCO
    Workshop on BBO Benchmarking 2012, which contains 21 state-of-the-
    art solvers run on 24 well-established functions.
    Original languageEnglish
    Title of host publicationLearning and Intelligent Optimization
    PublisherSpringer VS
    Publication dateJan 2013
    ISBN (Print)978-3-642-44972-7
    Publication statusPublished - Jan 2013
    Event Learning and Intelligent OptimizatioN Conference 2013 - Episcopate Museum Catania, Catania, Italy
    Duration: 7 Jan 201311 Jan 2013


    Conference Learning and Intelligent OptimizatioN Conference 2013
    LocationEpiscopate Museum Catania
    OtherEdited by Giuseppe Nicosia and Panos Pardalos
    Internet address
    SeriesLecture Notes in Computer Science


    • Black-box optimization
    • Fitness landscape analysis
    • Algorithm portfolio
    • Computationally intensive functions
    • Optimization benchmarking


    Dive into the research topics of 'Features for Exploiting Black-Box Optimization Problem Structure.'. Together they form a unique fingerprint.

    Cite this