Accuracy, parsimony, and generality in evolutionary learning systems via multiobjective selection

Xavier Llorà, David E. Goldberg, Ivan Traus, Ester Bernadó

    Research output: Book chapterConference contributionpeer-review

    17 Citations (Scopus)

    Abstract

    Evolutionary learning systems (also known as Pittsburgh learning classifier systems) need to balance accuracy and parsimony for evolving high quality general hypotheses. The learning process used in evolutionary learning systems is based on a set of training instances that sample the target concept to be learned. Thus, the learning process may overfit the learned hypothesis to the given set of training instances. In order to address some of these issues, this paper introduces a multiobjective approach to evolutionary learning systems. Thus, we translate the selection of promising hypotheses into a two-objective problem that looks for: (1) accurate (low error), and (2) compact (low complexity) solutions. Using the proposed multiobjective approach a set of compromise hypotheses are spread along the Pareto front. We also introduce a theory of the impact of noise when sampling the target concept to be learned, as well as the appearance of overfitted hypotheses as the result of perturbations on high quality generalization hypotheses in the Pareto front.

    Original languageEnglish
    Title of host publicationLearning Classifier Systems - 5th International Workshop, IWLCS 2002, Revised Papers
    EditorsPier Luca Lanzi, Wolfgang Stolzmann, Stewart W. Wilson, Stewart W. Wilson
    PublisherSpringer Verlag
    Pages118-142
    Number of pages25
    ISBN (Print)3540205446, 9783540205449
    DOIs
    Publication statusPublished - 2003
    Event5th International Workshop on Learning Classifier Systems, IWLCS 2002 - Granada, Spain
    Duration: 7 Sept 20028 Sept 2002

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume2661
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference5th International Workshop on Learning Classifier Systems, IWLCS 2002
    Country/TerritorySpain
    CityGranada
    Period7/09/028/09/02

    Fingerprint

    Dive into the research topics of 'Accuracy, parsimony, and generality in evolutionary learning systems via multiobjective selection'. Together they form a unique fingerprint.

    Cite this