Steady State Genetic Programming with Constrained Complexity Crossover Using Species Sub Population

Created by W.Langdon from gp-bibliography.bib Revision:1.8051

@InProceedings{watson:1997:ssGPcccssp,

• author = "Andrew H. Watson and Ian C. Parmee",
• title = "Steady State Genetic Programming with Constrained Complexity Crossover Using Species Sub Population",
• booktitle = "Genetic Algorithms: Proceedings of the Seventh International Conference",
• year = "1997",
• editor = "Thomas Back",
• pages = "315--321",
• address = "Michigan State University, East Lansing, MI, USA",
• publisher_address = "San Francisco, CA, USA",
• month = "19-23 " # jul,
• publisher = "Morgan Kaufmann",
• keywords = "genetic algorithms, genetic programming",
• ISBN = "1-55860-487-1",
• URL = "http://www.cs.ucl.ac.uk/staff/W.Langdon/ftp/papers/icga1997/Watson_1997_ssGPccc.pdf",
• size = "7 pages",
• abstract = "We introduce an alternative approach to Genetic Programming (GP), which is based upon a steady state population and a novel constrained complexity crossover operator. This technique, called {"}DRAM-GP{"} (i.e. Distributed, Rapid, Attenuated Memory Genetic Programming), uses node complexity weightings as a basis for speciation. The population is decomposed into smaller sub-populations which communicate with each other through the action of crossover. The effectiveness of this method is demonstrated by successful application to Boolean concept formation and to symbolic regression problems. The results show that improved performance is possible with a dramatic reduction in population size and associated memory requirements.",
• notes = "ICGA-97",

}

Genetic Programming entries for Andrew H Watson Ian C Parmee

Citations