Exploring Overfitting in Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.4420

  author =       "Gregory Paris and Denis Robilliard and Cyril Fonlupt",
  title =        "Exploring Overfitting in Genetic Programming",
  booktitle =    "Evolution Artificielle, 6th International Conference",
  year =         "2003",
  editor =       "Pierre Liardet and Pierre Collet and Cyril Fonlupt and 
                 Evelyne Lutton and Marc Schoenauer",
  volume =       "2936",
  series =       "Lecture Notes in Computer Science",
  pages =        "267--277",
  address =      "Marseilles, France",
  month =        "27-30 " # oct,
  publisher =    "Springer",
  note =         "Revised Selected Papers",
  keywords =     "genetic algorithms, genetic programming, Artificial
  ISBN =         "3-540-21523-9",
  DOI =          "doi:10.1007/b96080",
  abstract =     "The problem of overfitting (focusing closely on
                 examples at the loss of generalisation power) is
                 encountered in all supervised machine learning schemes.
                 This study is dedicated to explore some aspects of over
                 fitting in the particular case of genetic programming.
                 After recalling the causes usually invoked to explain
                 over-fitting such as hypothesis complexity or noisy
                 learning examples, we test and compare the resistance
                 to over fitting on three variants of genetic
                 programming algorithms (basic GP, sizefair crossover GP
                 and GP with boosting) on two benchmarks, a symbolic
                 regression and a classification problem. We propose
                 guidelines based on these results to help reduce over
                 fitting with genetic programming.",
  bibsource =    "DBLP, http://dblp.uni-trier.de",
  notes =        "EA'03

                 size fair crossover \cite{langdon:2000:fairxo}

Genetic Programming entries for Gregory Paris Denis Robilliard Cyril Fonlupt