Controlling Overfitting in Symbolic Regression Based on a Bias/Variance Error Decomposition

Created by W.Langdon from gp-bibliography.bib Revision:1.7615

@InProceedings{conf/ppsn/Agapitos12,

• author = "Alexandros Agapitos and Anthony Brabazon and Michael O'Neill",
• title = "Controlling Overfitting in Symbolic Regression Based on a Bias/Variance Error Decomposition",
• booktitle = "Parallel Problem Solving from Nature, PPSN XII (part 1)",
• year = "2012",
• editor = "Carlos A. {Coello Coello} and Vincenzo Cutello and Kalyanmoy Deb and Stephanie Forrest and Giuseppe Nicosia and Mario Pavone",
• volume = "7491",
• series = "Lecture Notes in Computer Science",
• pages = "438--447",
• address = "Taormina, Italy",
• month = sep # " 1-5",
• publisher = "Springer",
• keywords = "genetic algorithms, genetic programming",
• isbn13 = "978-3-642-32936-4",
• DOI = "doi:10.1007/978-3-642-32937-1_44",
• size = "10 pages",
• abstract = "We consider the fundamental property of generalisation of data-driven models evolved by means of Genetic Programming (GP). The statistical treatment of decomposing the regression error into bias and variance terms provides insight into the generalisation capability of this modelling method. The error decomposition is used as a source of inspiration to design a fitness function that relaxes the sensitivity of an evolved model to a particular training dataset. Results on eight symbolic regression problems show that new method is capable on inducing better-generalising models than standard GP for most of the problems.",
• affiliation = "Natural Computing Research and Applications Group, University College Dublin, Ireland",

}

Genetic Programming entries for Alexandros Agapitos Anthony Brabazon Michael O'Neill

Citations