Epsilon-lexicase Selection for Regression

Created by W.Langdon from gp-bibliography.bib Revision:1.4549

  author =       "William {La Cava} and Lee Spector and Kourosh Danai",
  title =        "Epsilon-lexicase Selection for Regression",
  booktitle =    "GECCO '16: Proceedings of the 2016 Annual Conference
                 on Genetic and Evolutionary Computation",
  year =         "2016",
  editor =       "Tobias Friedrich",
  pages =        "741--748",
  keywords =     "genetic algorithms, genetic programming",
  month =        "20-24 " # jul,
  organisation = "SIGEVO",
  address =      "Denver, USA",
  publisher =    "ACM",
  publisher_address = "New York, NY, USA",
  isbn13 =       "978-1-4503-4206-3",
  DOI =          "doi:10.1145/2908812.2908898",
  abstract =     "Lexicase selection is a parent selection method that
                 considers test cases separately, rather than in
                 aggregate, when performing parent selection. It
                 performs well in discrete error spaces but not on the
                 continuous-valued problems that compose most system
                 identification tasks. In this paper, we develop a new
                 form of lexicase selection for symbolic regression,
                 named epsilon-lexicase selection, that redefines the
                 pass condition for individuals on each test case in a
                 more effective way. We run a series of experiments on
                 real-world and synthetic problems with several
                 treatments of epsilon and quantify how e affects parent
                 selection and model performance. e-lexicase selection
                 is shown to be effective for regression, producing
                 better fit models compared to other techniques such as
                 tournament selection and age-fitness Pareto
                 optimization. We demonstrate that epsilon can be
                 adapted automatically for individual test cases based
                 on the population performance distribution. Our
                 experiments show that e-lexicase selection with
                 automatic epsilon produces the most accurate models
                 across tested problems with negligible computational
                 overhead. We show that behavioural diversity is
                 exceptionally high in lexicase selection treatments,
                 and that e-lexicase selection makes use of more fitness
                 cases when selecting parents than lexicase selection,
                 which helps explain the performance improvement.",
  notes =        "University of Massachusetts Amherst, Hampshire

                 GECCO-2016 A Recombination of the 25th International
                 Conference on Genetic Algorithms (ICGA-2016) and the
                 21st Annual Genetic Programming Conference (GP-2016)",

Genetic Programming entries for William La Cava Lee Spector Kourosh Danai