Genetic network programming with automatically generated macro nodes of variable size

Shingo Mabu, Hiroyuki Hatakeyama, Hiroshi Nakagoe, Kotaro Hirasawa, Takayuki Furuzuki

Research output: Contribution to journalArticle

Abstract

Recently, Genetic Network Programming (GNP) has been proposed as one of the evolutionary algorithms. It represents its solutions as directed graph structures and the distinguished abilities have been shown. However, when we apply GNP to complex problems like the real world one, GNP must have robustness against the changes of environments and evolve quickly. Therefore, we introduced Automatically Generated Macro Nodes (AGMs) to GNP (GNP with AGMs). Actually GNP with AGMs has shown higher performances than the conventional GNP in terms of the fitness and the speed of evolution. In this paper, a new mechanism, AGMs with variable size, is introduced to GNP. Conventional AGMs have the fixed number of nodes and they evolve using only genetic operations, while a new method allows AGM to add nodes by necessity and delete nodes which do not contribute to the evolution of the AGM. The proposed GNP with AGMs of variable size is expected to evolve effectively and efficiently when it is applied to agent systems and also expected to make better behavior sequences of agents more easily than the conventional GNP algorithm. In the simulations, the proposed and conventional methods are applied to a tileworld problem and they are compared with each other. From the results, GNP with AGMs of variable size shows better fitness than GNP with AGMs of fixed size and the conventional GNP when adapting ten different environments.

Original languageEnglish
JournalIEEJ Transactions on Electronics, Information and Systems
Volume126
Issue number4
Publication statusPublished - 2006 Jan 1

Fingerprint

Computer programming
Macros
Genetic programming
Directed graphs
Evolutionary algorithms

Keywords

  • Automatically Defined Function (ADF)
  • Evolutionary Computation
  • Genetic Network Programming
  • Macro Node with Variable Size
  • Tileworld

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Genetic network programming with automatically generated macro nodes of variable size. / Mabu, Shingo; Hatakeyama, Hiroyuki; Nakagoe, Hiroshi; Hirasawa, Kotaro; Furuzuki, Takayuki.

In: IEEJ Transactions on Electronics, Information and Systems, Vol. 126, No. 4, 01.01.2006.

Research output: Contribution to journalArticle

@article{a3d997cc32f34bb89d92e19c4a93fdba,
title = "Genetic network programming with automatically generated macro nodes of variable size",
abstract = "Recently, Genetic Network Programming (GNP) has been proposed as one of the evolutionary algorithms. It represents its solutions as directed graph structures and the distinguished abilities have been shown. However, when we apply GNP to complex problems like the real world one, GNP must have robustness against the changes of environments and evolve quickly. Therefore, we introduced Automatically Generated Macro Nodes (AGMs) to GNP (GNP with AGMs). Actually GNP with AGMs has shown higher performances than the conventional GNP in terms of the fitness and the speed of evolution. In this paper, a new mechanism, AGMs with variable size, is introduced to GNP. Conventional AGMs have the fixed number of nodes and they evolve using only genetic operations, while a new method allows AGM to add nodes by necessity and delete nodes which do not contribute to the evolution of the AGM. The proposed GNP with AGMs of variable size is expected to evolve effectively and efficiently when it is applied to agent systems and also expected to make better behavior sequences of agents more easily than the conventional GNP algorithm. In the simulations, the proposed and conventional methods are applied to a tileworld problem and they are compared with each other. From the results, GNP with AGMs of variable size shows better fitness than GNP with AGMs of fixed size and the conventional GNP when adapting ten different environments.",
keywords = "Automatically Defined Function (ADF), Evolutionary Computation, Genetic Network Programming, Macro Node with Variable Size, Tileworld",
author = "Shingo Mabu and Hiroyuki Hatakeyama and Hiroshi Nakagoe and Kotaro Hirasawa and Takayuki Furuzuki",
year = "2006",
month = "1",
day = "1",
language = "English",
volume = "126",
journal = "IEEJ Transactions on Electronics, Information and Systems",
issn = "0385-4221",
publisher = "The Institute of Electrical Engineers of Japan",
number = "4",

}

TY - JOUR

T1 - Genetic network programming with automatically generated macro nodes of variable size

AU - Mabu, Shingo

AU - Hatakeyama, Hiroyuki

AU - Nakagoe, Hiroshi

AU - Hirasawa, Kotaro

AU - Furuzuki, Takayuki

PY - 2006/1/1

Y1 - 2006/1/1

N2 - Recently, Genetic Network Programming (GNP) has been proposed as one of the evolutionary algorithms. It represents its solutions as directed graph structures and the distinguished abilities have been shown. However, when we apply GNP to complex problems like the real world one, GNP must have robustness against the changes of environments and evolve quickly. Therefore, we introduced Automatically Generated Macro Nodes (AGMs) to GNP (GNP with AGMs). Actually GNP with AGMs has shown higher performances than the conventional GNP in terms of the fitness and the speed of evolution. In this paper, a new mechanism, AGMs with variable size, is introduced to GNP. Conventional AGMs have the fixed number of nodes and they evolve using only genetic operations, while a new method allows AGM to add nodes by necessity and delete nodes which do not contribute to the evolution of the AGM. The proposed GNP with AGMs of variable size is expected to evolve effectively and efficiently when it is applied to agent systems and also expected to make better behavior sequences of agents more easily than the conventional GNP algorithm. In the simulations, the proposed and conventional methods are applied to a tileworld problem and they are compared with each other. From the results, GNP with AGMs of variable size shows better fitness than GNP with AGMs of fixed size and the conventional GNP when adapting ten different environments.

AB - Recently, Genetic Network Programming (GNP) has been proposed as one of the evolutionary algorithms. It represents its solutions as directed graph structures and the distinguished abilities have been shown. However, when we apply GNP to complex problems like the real world one, GNP must have robustness against the changes of environments and evolve quickly. Therefore, we introduced Automatically Generated Macro Nodes (AGMs) to GNP (GNP with AGMs). Actually GNP with AGMs has shown higher performances than the conventional GNP in terms of the fitness and the speed of evolution. In this paper, a new mechanism, AGMs with variable size, is introduced to GNP. Conventional AGMs have the fixed number of nodes and they evolve using only genetic operations, while a new method allows AGM to add nodes by necessity and delete nodes which do not contribute to the evolution of the AGM. The proposed GNP with AGMs of variable size is expected to evolve effectively and efficiently when it is applied to agent systems and also expected to make better behavior sequences of agents more easily than the conventional GNP algorithm. In the simulations, the proposed and conventional methods are applied to a tileworld problem and they are compared with each other. From the results, GNP with AGMs of variable size shows better fitness than GNP with AGMs of fixed size and the conventional GNP when adapting ten different environments.

KW - Automatically Defined Function (ADF)

KW - Evolutionary Computation

KW - Genetic Network Programming

KW - Macro Node with Variable Size

KW - Tileworld

UR - http://www.scopus.com/inward/record.url?scp=33645575550&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645575550&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:33645575550

VL - 126

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

SN - 0385-4221

IS - 4

ER -