- author = "David Corney and Ian Parmee",
- title = "N-Dimensional Surface Mapping Using Genetic Programming",
- booktitle = "Proceedings of the Genetic and Evolutionary Computation Conference",
- year = "1999",
- editor = "Wolfgang Banzhaf and Jason Daida and Agoston E. Eiben and Max H. Garzon and Vasant Honavar and Mark Jakiela and Robert E. Smith",
- volume = "2",
- pages = "1230",
- address = "Orlando, Florida, USA",
- publisher_address = "San Francisco, CA 94104, USA",
- month = "13-17 " # jul,
- publisher = "Morgan Kaufmann",
- keywords = "genetic algorithms, genetic programming, poster papers",
- ISBN = "1-55860-611-4",
-
URL = "
http://gpbib.cs.ucl.ac.uk/gecco1999/GP-424.pdf",
-
URL = "
http://gpbib.cs.ucl.ac.uk/gecco1999/GP-424.ps",
-
URL = "http://dcorney.com/papers/gecco_surfaceGP.pdf",
- size = "1 page",
- abstract = "This work introduces an extension to Genetic Programming (GP), known as GP-UDF which uses multiple User-Defined Functions (UDFs) to solve surface mapping problems. UDFs are high level primitives, such as polynomials and Gaussian hills, which simplify mapping and aid human interpretation of GP results. Preliminary results show that although UDFs do not improve GP accuracy, they may aid in landscape classification.",
-
notes = "GECCO-99 A joint meeting of the eighth international
conference on genetic algorithms (ICGA-99) and the
fourth annual genetic programming conference
(GP-99)
MSc Thesis -- text from http://www.cs.ucl.ac.uk/staff/D.Corney/MSc_thesis_abstract.html
N-Dimensional Surface Mapping Using Genetic Programming
This work introduces an extension to Genetic Programming (GP) known as {"}GP-UDF{"} which uses multiple User-Defined Functions (UDFs) to solve surface-mapping problems. These UDFs are high-level primitives, such as hills and polynomials, which compress the information required to map a surface. UDFs can be used to add real-world knowledge to a genetic search, and also to analyse and classify high-dimensional surfaces. GP-UDF also produces more readable solutions than standard GP.
The results show that, for the problems considered, GP-UDF does not produce more accurate models than standard GP. However, the results also suggest that GP-UDF could be used as a {"}landscape classifier{"}, a tool for analysing high-dimensional surfaces to identify characteristic features.
An important consideration in systems identification is the transparency (i.e. readability), of a model. GP-UDF is compared with neural networks (both MLP and RBF networks), and is shown to be far more readable, with the cost of being less accurate.",
