A Linear Regression Approach to Numerical Simplification in Tree-Based Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.7630

@TechReport{Johnston:tr09-7,

• author = "Mark Johnston and Thomas Liddle and Mengjie Zhang",
• title = "A Linear Regression Approach to Numerical Simplification in Tree-Based Genetic Programming",
• institution = "School of Mathematics Statistics and Operations Research, Victoria University of Wellington",
• year = "2009",
• type = "Research report",
• number = "09-7",
• address = "New Zealand",
• month = "14 " # dec,
• keywords = "genetic algorithms, genetic programming",
• URL = "http://msor.victoria.ac.nz/twiki/pub/Main/ResearchReportSeries/msor09-07.pdf",
• abstract = "We propose a novel approach to simplification in tree-based Genetic Programming to combat program bloat, based upon numerical relaxations of algebraic rules.We also separate proposal of simplifications (using linear regression, removing redundant children, and replacing small ranges with a constant) from an acceptance criterion that checks the effect of proposed simplifications on the evaluation of training examples, looking several levels up the tree.We test our simplification method on three classification datasets and conclude that the success of linear regression is data set dependent, that looking further up the tree can catch unwanted bad case simplifications, and that CPU time can be significantly reduced while maintaining classification accuracy on unseen examples.",
• notes = "Wine, Wisconsin, Coins",
• size = "38 pages",

}

Genetic Programming entries for Mark Johnston Thomas Liddle Mengjie Zhang

Citations