Incorporating expert knowledge in evolutionary search: a study of seeding methods

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

@InProceedings{DBLP:conf/gecco/SchmidtL09a,
  author =       "Michael D. Schmidt and Hod Lipson",
  title =        "Incorporating expert knowledge in evolutionary search:
                 a study of seeding methods",
  booktitle =    "GECCO '09: Proceedings of the 11th Annual conference
                 on Genetic and evolutionary computation",
  year =         "2009",
  editor =       "Guenther Raidl and Franz Rothlauf and 
                 Giovanni Squillero and Rolf Drechsler and Thomas Stuetzle and 
                 Mauro Birattari and Clare Bates Congdon and 
                 Martin Middendorf and Christian Blum and Carlos Cotta and 
                 Peter Bosman and Joern Grahl and Joshua Knowles and 
                 David Corne and Hans-Georg Beyer and Ken Stanley and 
                 Julian F. Miller and Jano {van Hemert} and 
                 Tom Lenaerts and Marc Ebner and Jaume Bacardit and 
                 Michael O'Neill and Massimiliano {Di Penta} and Benjamin Doerr and 
                 Thomas Jansen and Riccardo Poli and Enrique Alba",
  pages =        "1091--1098",
  address =      "Montreal",
  publisher =    "ACM",
  publisher_address = "New York, NY, USA",
  month =        "8-12 " # jul,
  organisation = "SigEvo",
  keywords =     "genetic algorithms, genetic programming",
  isbn13 =       "978-1-60558-325-9",
  bibsource =    "DBLP, http://dblp.uni-trier.de",
  DOI =          "doi:10.1145/1569901.1570048",
  abstract =     "We investigated several methods for using expert
                 knowledge in evolutionary search, and compared their
                 impact on performance and scalability into increasingly
                 complex problems. We collected data over one thousand
                 randomly generated problems. We then simulated
                 collecting expert knowledge for each problem by
                 optimizing an approximated version of the exact
                 solution. We then compared six different methods of
                 seeding the approximate model in to the genetic
                 program, such as using the entire approximate model at
                 once or breaking it into pieces. Contrary to common
                 intuition, we found that inserting the complete expert
                 solution into the population is not the best way to use
                 that information; using parts of that solution is often
                 more effective. Additionally, we found that each method
                 scaled differently based on the complexity and accuracy
                 of the approximate solution. Inserting randomized
                 pieces of the approximate solution into the population
                 scaled the best into high complexity problems and was
                 the most invariant to the accuracy of the approximate
                 solution. Furthermore, this method produced the least
                 bloated solutions of all methods. In general, methods
                 that used randomized parameter coefficients scaled best
                 with the approximate error, and methods that inserted
                 entire approximate solutions scaled worst with the
                 problem complexity.",
  notes =        "GECCO-2009 A joint meeting of the eighteenth
                 international conference on genetic algorithms
                 (ICGA-2009) and the fourteenth annual genetic
                 programming conference (GP-2009).

                 ACM Order Number 910092.",
}

Genetic Programming entries for Michael D Schmidt Hod Lipson

Citations