Publication detail

Efficient Phenotype Evaluation in Cartesian Genetic Programming

VAŠÍČEK, Z. SLANÝ, K.

Original Title

Efficient Phenotype Evaluation in Cartesian Genetic Programming

English Title

Efficient Phenotype Evaluation in Cartesian Genetic Programming

Type

conference paper

Language

en

Original Abstract

This paper describes an efficient acceleration technique designed to speedup the evaluation of candidate solutions in Cartesian Genetic Programming (CGP). The method is based on translation of the CGP phenotype to a binary machine code that is consequently executed. The key feature of the presented approach is that the introduction of the translation mechanism into common fitness evaluation procedure requires only marginal knowledge of target CPU instruction set. The proposed acceleration technique is evaluated using a symbolic regression problem in floating point domain. It is shown that for a cost of small changes in a common CGP implementation, a significant speedup can be obtained even on a common desktop CPU.

English abstract

This paper describes an efficient acceleration technique designed to speedup the evaluation of candidate solutions in Cartesian Genetic Programming (CGP). The method is based on translation of the CGP phenotype to a binary machine code that is consequently executed. The key feature of the presented approach is that the introduction of the translation mechanism into common fitness evaluation procedure requires only marginal knowledge of target CPU instruction set. The proposed acceleration technique is evaluated using a symbolic regression problem in floating point domain. It is shown that for a cost of small changes in a common CGP implementation, a significant speedup can be obtained even on a common desktop CPU.

Keywords

genetic algorithms, genetic programming, cartesian genetic programming, Fitness evaluation, Acceleration, Symbolic regression

RIV year

2012

Released

25.01.2012

Publisher

Springer Verlag

Location

Heidelberg

ISBN

978-3-642-29138-8

Book

Proc. of the 15th European Conference on Genetic Programming

Edition

Lecture Notes in Computer Science

Edition number

NEUVEDEN

Pages from

266

Pages to

278

Pages count

12

Documents

BibTex


@inproceedings{BUT96987,
  author="Zdeněk {Vašíček} and Karel {Slaný}",
  title="Efficient Phenotype Evaluation in Cartesian Genetic Programming",
  annote="This paper describes an efficient acceleration technique designed to speedup the
evaluation of candidate solutions in Cartesian Genetic Programming (CGP). The
method is based on translation of the CGP phenotype to a binary machine code that
is consequently executed. The key feature of the presented approach is that the
introduction of the translation mechanism into common fitness evaluation
procedure requires only marginal knowledge of target CPU instruction set. The
proposed acceleration technique is evaluated using a symbolic regression problem
in floating point domain. It is shown that for a cost of small changes in a
common CGP implementation, a significant speedup can be obtained even on a common
desktop CPU.",
  address="Springer Verlag",
  booktitle="Proc. of the 15th European Conference on Genetic Programming",
  chapter="96987",
  doi="10.1007/978-3-642-29139-5_23",
  edition="Lecture Notes in Computer Science",
  howpublished="online",
  institution="Springer Verlag",
  year="2012",
  month="january",
  pages="266--278",
  publisher="Springer Verlag",
  type="conference paper"
}