MT-CGP: mixed type cartesian genetic programming

Simon Harding; Vincent Graziano; Jürgen Leitner; Jürgen Schmidhuber

doi:10.1145/2330163.2330268

MT-CGP: mixed type cartesian genetic programming

Simon Harding, Vincent Graziano, Jürgen Leitner, Jürgen Schmidhuber

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

33 Citations (Scopus)

Abstract

The majority of genetic programming implementations build expressions that only use a single data type. This is in contrast to human engineered programs that typically make use of multiple data types, as this provides the ability to express solutions in a more natural fashion. In this paper, we present a version of Cartesian Genetic Programming that handles multiple data types. We demonstrate that this allows evolution to quickly find competitive, compact, and human readable solutions on multiple classification tasks.

Original language	English
Title of host publication	GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	751-758
Number of pages	8
ISBN (Print)	9781450311779
DOIs	https://doi.org/10.1145/2330163.2330268
Publication status	Published - 2012
Externally published	Yes
Event	The Genetic and Evolutionary Computation Conference 2012 - Philadelphia, United States of America Duration: 7 Jul 2012 → 11 Jul 2012 Conference number: 14th https://dl.acm.org/doi/proceedings/10.1145/2330784 (Proceedings)

Conference

Conference	The Genetic and Evolutionary Computation Conference 2012
Abbreviated title	GECCO 2012
Country/Territory	United States of America
City	Philadelphia
Period	7/07/12 → 11/07/12
Internet address	https://dl.acm.org/doi/proceedings/10.1145/2330784 (Proceedings)

Keywords

cartesian genetic programming
classifiers

Access to Document

10.1145/2330163.2330268

Cite this

@inproceedings{cdbbf21d46c84aae9526aa4acbae0764,

title = "MT-CGP: mixed type cartesian genetic programming",

abstract = "The majority of genetic programming implementations build expressions that only use a single data type. This is in contrast to human engineered programs that typically make use of multiple data types, as this provides the ability to express solutions in a more natural fashion. In this paper, we present a version of Cartesian Genetic Programming that handles multiple data types. We demonstrate that this allows evolution to quickly find competitive, compact, and human readable solutions on multiple classification tasks.",

keywords = "cartesian genetic programming, classifiers",

author = "Simon Harding and Vincent Graziano and J{\"u}rgen Leitner and J{\"u}rgen Schmidhuber",

year = "2012",

doi = "10.1145/2330163.2330268",

language = "English",

isbn = "9781450311779",

pages = "751--758",

booktitle = "GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "The Genetic and Evolutionary Computation Conference 2012, GECCO 2012 ; Conference date: 07-07-2012 Through 11-07-2012",

url = "https://dl.acm.org/doi/proceedings/10.1145/2330784",

}

Harding, S, Graziano, V, Leitner, J & Schmidhuber, J 2012, MT-CGP: mixed type cartesian genetic programming. in GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation. IEEE, Institute of Electrical and Electronics Engineers, pp. 751-758, The Genetic and Evolutionary Computation Conference 2012, Philadelphia, Pennsylvania, United States of America, 7/07/12. https://doi.org/10.1145/2330163.2330268

MT-CGP: mixed type cartesian genetic programming. / Harding, Simon; Graziano, Vincent; Leitner, Jürgen et al.
GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation. IEEE, Institute of Electrical and Electronics Engineers, 2012. p. 751-758.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - MT-CGP

T2 - The Genetic and Evolutionary Computation Conference 2012

AU - Harding, Simon

AU - Graziano, Vincent

AU - Leitner, Jürgen

AU - Schmidhuber, Jürgen

N1 - Conference code: 14th

PY - 2012

Y1 - 2012

N2 - The majority of genetic programming implementations build expressions that only use a single data type. This is in contrast to human engineered programs that typically make use of multiple data types, as this provides the ability to express solutions in a more natural fashion. In this paper, we present a version of Cartesian Genetic Programming that handles multiple data types. We demonstrate that this allows evolution to quickly find competitive, compact, and human readable solutions on multiple classification tasks.

AB - The majority of genetic programming implementations build expressions that only use a single data type. This is in contrast to human engineered programs that typically make use of multiple data types, as this provides the ability to express solutions in a more natural fashion. In this paper, we present a version of Cartesian Genetic Programming that handles multiple data types. We demonstrate that this allows evolution to quickly find competitive, compact, and human readable solutions on multiple classification tasks.

KW - cartesian genetic programming

KW - classifiers

UR - http://www.scopus.com/inward/record.url?scp=84864645762&partnerID=8YFLogxK

U2 - 10.1145/2330163.2330268

DO - 10.1145/2330163.2330268

M3 - Conference Paper

AN - SCOPUS:84864645762

SN - 9781450311779

SP - 751

EP - 758

BT - GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation

PB - IEEE, Institute of Electrical and Electronics Engineers

Y2 - 7 July 2012 through 11 July 2012

ER -

MT-CGP: mixed type cartesian genetic programming

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this