Variable Size Genetic Network Programming

Katagiri Hironobu, Takayuki Furuzuki, Murata Junichi, Hirasawa Kotaro

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Genetic Network Programming (GNP) is a kind of evolutionary methods, which evolves arbitrary directed graph programs. Previously, the program size of GNP was fixed. In the paper, a new method is proposed, where the program size is adaptively changed depending on the frequency of the use of nodes. To control and to decide a program size are important and difficult problems in Evolutionary Computation, especially, a well-known crossover operator tends to cause bloat. We introduce two additional operators, add operator and delete operator, that can change the number of each kind of nodes based on whether a node function is important in the environment or not. Simulation results shows that the proposed method brings about extremely better results compared with ordinary fixed size GNP.

Original languageEnglish
Pages (from-to)57-66
Number of pages10
JournalIEEJ Transactions on Electronics, Information and Systems
Volume123
Issue number1
DOIs
Publication statusPublished - 2003
Externally publishedYes

Fingerprint

Directed graphs
Evolutionary algorithms
Mathematical operators

Keywords

  • arbitrary directed graph
  • Evolutionary Computation
  • Genetic Network Programming
  • Genetic Programming
  • planning
  • the tileworld

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Variable Size Genetic Network Programming. / Hironobu, Katagiri; Furuzuki, Takayuki; Junichi, Murata; Kotaro, Hirasawa.

In: IEEJ Transactions on Electronics, Information and Systems, Vol. 123, No. 1, 2003, p. 57-66.

Research output: Contribution to journalArticle

Hironobu, Katagiri ; Furuzuki, Takayuki ; Junichi, Murata ; Kotaro, Hirasawa. / Variable Size Genetic Network Programming. In: IEEJ Transactions on Electronics, Information and Systems. 2003 ; Vol. 123, No. 1. pp. 57-66.
@article{b1bf9a4e80a743dc9bc4ec0caef48137,
title = "Variable Size Genetic Network Programming",
abstract = "Genetic Network Programming (GNP) is a kind of evolutionary methods, which evolves arbitrary directed graph programs. Previously, the program size of GNP was fixed. In the paper, a new method is proposed, where the program size is adaptively changed depending on the frequency of the use of nodes. To control and to decide a program size are important and difficult problems in Evolutionary Computation, especially, a well-known crossover operator tends to cause bloat. We introduce two additional operators, add operator and delete operator, that can change the number of each kind of nodes based on whether a node function is important in the environment or not. Simulation results shows that the proposed method brings about extremely better results compared with ordinary fixed size GNP.",
keywords = "arbitrary directed graph, Evolutionary Computation, Genetic Network Programming, Genetic Programming, planning, the tileworld",
author = "Katagiri Hironobu and Takayuki Furuzuki and Murata Junichi and Hirasawa Kotaro",
year = "2003",
doi = "10.1541/ieejeiss.123.57",
language = "English",
volume = "123",
pages = "57--66",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "1",

}

TY - JOUR

T1 - Variable Size Genetic Network Programming

AU - Hironobu, Katagiri

AU - Furuzuki, Takayuki

AU - Junichi, Murata

AU - Kotaro, Hirasawa

PY - 2003

Y1 - 2003

N2 - Genetic Network Programming (GNP) is a kind of evolutionary methods, which evolves arbitrary directed graph programs. Previously, the program size of GNP was fixed. In the paper, a new method is proposed, where the program size is adaptively changed depending on the frequency of the use of nodes. To control and to decide a program size are important and difficult problems in Evolutionary Computation, especially, a well-known crossover operator tends to cause bloat. We introduce two additional operators, add operator and delete operator, that can change the number of each kind of nodes based on whether a node function is important in the environment or not. Simulation results shows that the proposed method brings about extremely better results compared with ordinary fixed size GNP.

AB - Genetic Network Programming (GNP) is a kind of evolutionary methods, which evolves arbitrary directed graph programs. Previously, the program size of GNP was fixed. In the paper, a new method is proposed, where the program size is adaptively changed depending on the frequency of the use of nodes. To control and to decide a program size are important and difficult problems in Evolutionary Computation, especially, a well-known crossover operator tends to cause bloat. We introduce two additional operators, add operator and delete operator, that can change the number of each kind of nodes based on whether a node function is important in the environment or not. Simulation results shows that the proposed method brings about extremely better results compared with ordinary fixed size GNP.

KW - arbitrary directed graph

KW - Evolutionary Computation

KW - Genetic Network Programming

KW - Genetic Programming

KW - planning

KW - the tileworld

UR - http://www.scopus.com/inward/record.url?scp=4344646172&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=4344646172&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.123.57

DO - 10.1541/ieejeiss.123.57

M3 - Article

AN - SCOPUS:4344646172

VL - 123

SP - 57

EP - 66

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 1

ER -