Smooth Fitting with a Method for Determining the Regularization Parameter under the Genetic Programming Algorithm

Created by W.Langdon from gp-bibliography.bib Revision:1.3872

@Article{Yun01,
  author =       "Yun Seog Yeun and Kyung Ho Lee and Sang Min Han and 
                 Young Soon Yang",
  title =        "Smooth Fitting with a Method for Determining the
                 Regularization Parameter under the Genetic Programming
                 Algorithm",
  journal =      "Information Sciences",
  year =         "2001",
  volume =       "133",
  number =       "3-4",
  pages =        "175--194",
  month =        apr,
  email =        "yeonyun@road.daejin.ac.kr",
  keywords =     "genetic algorithms, genetic programming, Smooth
                 fitting, Regularization parameter",
  DOI =          "doi:10.1016/S0020-0255(01)00084-6",
  abstract =     "This paper deals with the smooth fitting problem under
                 the genetic programming(GP) algorithm. To reduce the
                 computational cost required for evaluating the fitness
                 value of GP trees, numerical weights of GP trees are
                 estimated by adopting both linear associative memories
                 and the Hook & Jeeves method. The quality of smooth
                 fitting is critically dependent on the choice of the
                 regularization parameter. So, we present a novel method
                 for choosing the regularization parameter. Two
                 numerical examples are given with the comparison of
                 generalized cross-validation B-splines",
  notes =        "Euclidean norm = zero-order Tikhonov regularisation,
                 is not sufficient p178 uses(?) weighting based on first
                 derivative of evolved function but too CPU
                 expensive(?). LAM HJ discrepancy principle DP
                 cross-validation CV composite residual and smoothing
                 operator CRESO L-curve zero crossing ZC considered in
                 text but use heuristic

                 two test functions 2*(sin(t))**4
                 1.5*(exp(-30*(t-0.25)**2)+sin(pi(t-0.2))**2)

                 {"}..sometimes the GP tree that is discarded by the
                 criterion proposed in this paper is far better than the
                 tree selected as the best one.{"} p192

                 Information Sciences
                 http://www.elsevier.com/inca/publications/store/5/0/5/7/3/0/505730.pub.htt",
}

Genetic Programming entries for Yun Seog Yeun Kyung Ho Lee Sang Min Han Young-Soon Yang

Citations