Optimal Experiment Design for Coevolutionary Active Learning

Created by W.Langdon from gp-bibliography.bib Revision:1.3973

@Article{Ly:2014:ieeeTEC,
  author =       "Daniel Le Ly and Hod Lipson",
  title =        "Optimal Experiment Design for Coevolutionary Active
                 Learning",
  journal =      "IEEE Transactions on Evolutionary Computation",
  year =         "2014",
  volume =       "18",
  number =       "3",
  pages =        "394--404",
  month =        jun,
  keywords =     "genetic algorithms, genetic programming, Active
                 Learning, Competitive Coevolution, Optimal Experiment
                 Design, Shannon Information Criterion",
  ISSN =         "1089-778X",
  DOI =          "doi:10.1109/TEVC.2013.2281529",
  size =         "11 pages",
  abstract =     "This paper presents a policy for selecting the most
                 informative individuals in a teacher-learner type
                 coevolution. We propose the use of the surprisal of the
                 mean, based on Shannon information theory, which best
                 disambiguates a collection of arbitrary and competing
                 models based solely on their predictions. This policy
                 is demonstrated within an iterative, coevolutionary
                 framework consisting of symbolic regression for model
                 inference and a genetic algorithm for optimal
                 experiment design. Complex, symbolic expressions are
                 reliably inferred using fewer than 32 observations. The
                 policy requires 21percent fewer experiments for model
                 inference compared to baselines and is particularly
                 effective in the presence of noise corruption, local
                 information content as well as high dimensional
                 systems. Furthermore, the policy was applied in a
                 real-world setting to model concrete compression
                 strength, where it was able to achieve 96.1percent of
                 the passive machine learning baseline performance with
                 only 16.6percent of the data.",
  notes =        "also known as \cite{6595614}",
}

Genetic Programming entries for Daniel L Ly Hod Lipson

Citations