Evolving Developmental Programs That Build Neural Networks for Solving Multiple Problems

Created by W.Langdon from gp-bibliography.bib Revision:1.8010

@InProceedings{miller:2018:GPTP,

• author = "Julian F. Miller and Dennis G. Wilson and Sylvain Cussat-Blanc",
• title = "Evolving Developmental Programs That Build Neural Networks for Solving Multiple Problems",
• booktitle = "Genetic Programming Theory and Practice XVI",
• year = "2018",
• editor = "Wolfgang Banzhaf and Lee Spector and Leigh Sheneman",
• pages = "137--178",
• address = "Ann Arbor, USA",
• month = "17-20 " # may,
• publisher = "Springer",
• keywords = "genetic algorithms, genetic programming, Cartesian Genetic Programming",
• isbn13 = "978-3-030-04734-4",
• URL = "http://link.springer.com/chapter/10.1007/978-3-030-04735-1_8",
• DOI = "doi:10.1007/978-3-030-04735-1_8",
• abstract = "A developmental model of an artificial neuron is presented. In this model, a pair of neural developmental programs develop an entire artificial neural network of arbitrary size. The pair of neural chromosomes are evolved using Cartesian Genetic Programming. During development, neurons and their connections can move, change, die or be created. We show that this two-chromosome genotype can be evolved to develop into a single neural network from which multiple conventional artificial neural networks can be extracted. The extracted conventional ANNs share some neurons across tasks. We have evaluated the performance of this method on three standard classification problems: cancer, diabetes and the glass datasets. The evolved pair of neuron programs can generate artificial neural networks that perform reasonably well on all three benchmark problems simultaneously. It appears to be the first attempt to solve multiple standard classification problems using a developmental approach.",

}

Genetic Programming entries for Julian F Miller Dennis G Wilson Sylvain Cussat-Blanc

Citations