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

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

    Producción científica: Capítulo del libroContribución a congreso/conferenciarevisión exhaustiva

    17 Citas (Scopus)

    Resumen

    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.

    Idioma originalInglés
    Título de la publicación alojadaLearning Classifier Systems - 5th International Workshop, IWLCS 2002, Revised Papers
    EditoresPier Luca Lanzi, Wolfgang Stolzmann, Stewart W. Wilson, Stewart W. Wilson
    EditorialSpringer Verlag
    Páginas118-142
    Número de páginas25
    ISBN (versión impresa)3540205446, 9783540205449
    DOI
    EstadoPublicada - 2003
    Evento5th International Workshop on Learning Classifier Systems, IWLCS 2002 - Granada, Espana
    Duración: 7 sept 20028 sept 2002

    Serie de la publicación

    NombreLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volumen2661
    ISSN (versión impresa)0302-9743
    ISSN (versión digital)1611-3349

    Conferencia

    Conferencia5th International Workshop on Learning Classifier Systems, IWLCS 2002
    País/TerritorioEspana
    CiudadGranada
    Período7/09/028/09/02

    Huella

    Profundice en los temas de investigación de 'Accuracy, parsimony, and generality in evolutionary learning systems via multiobjective selection'. En conjunto forman una huella única.

    Citar esto