Co-evolution of Hetero Multiagent Systems using Genetic Network Programming

Hirasawa Kotaro; Okubo Masafumi; Hu Jinglu; Murata Junichi; Matsuya Yuko

doi:10.1541/ieejeiss.123.544

Co-evolution of Hetero Multiagent Systems using Genetic Network Programming

Hirasawa Kotaro, Okubo Masafumi, Hu Jinglu, Murata Junichi, Matsuya Yuko

Research output: Contribution to journal › Article › peer-review

2 Citations (Scopus)

Abstract

Recently, many methods of evolutionary computation such as Genetic Algorithm(GA) and Genetic Programming(GP) have been developed as a basic tool for modeling and optimizing complex systems. GA has the genome of string structure, while the genome in GP is of tree structure. In this paper, a new evolutionary method named Genetic Network Programming(GNP), whose genome has network structure is applied to multiagent sysytems. Hetero Multiagent Sysytems with GNP are studied, where each agent of the same group has its own GNP program in order to build the adaptive agents against changing environments. Specifically, the comparison between Hetero Multiagent Systems and conventional Homo Multiagent Sysytems is carried out in simulations on ants behaviors.

Original language	English
Pages (from-to)	544-551
Number of pages	8
Journal	IEEJ Transactions on Electronics, Information and Systems
Volume	123
Issue number	3
DOIs	https://doi.org/10.1541/ieejeiss.123.544
Publication status	Published - 2003 Jan
Externally published	Yes

Keywords

GA
GNP
GP
artificial life
emergent systems
multi agent system

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1541/ieejeiss.123.544

Cite this

@article{7e14effa7d1b4f3e8104a7686a974ff8,

title = "Co-evolution of Hetero Multiagent Systems using Genetic Network Programming",

abstract = "Recently, many methods of evolutionary computation such as Genetic Algorithm(GA) and Genetic Programming(GP) have been developed as a basic tool for modeling and optimizing complex systems. GA has the genome of string structure, while the genome in GP is of tree structure. In this paper, a new evolutionary method named Genetic Network Programming(GNP), whose genome has network structure is applied to multiagent sysytems. Hetero Multiagent Sysytems with GNP are studied, where each agent of the same group has its own GNP program in order to build the adaptive agents against changing environments. Specifically, the comparison between Hetero Multiagent Systems and conventional Homo Multiagent Sysytems is carried out in simulations on ants behaviors.",

keywords = "GA, GNP, GP, artificial life, emergent systems, multi agent system",

author = "Hirasawa Kotaro and Okubo Masafumi and Hu Jinglu and Murata Junichi and Matsuya Yuko",

year = "2003",

month = jan,

doi = "10.1541/ieejeiss.123.544",

language = "English",

volume = "123",

pages = "544--551",

journal = "IEEJ Transactions on Electronics, Information and Systems",

issn = "0385-4221",

publisher = "The Institute of Electrical Engineers of Japan",

number = "3",

}

TY - JOUR

T1 - Co-evolution of Hetero Multiagent Systems using Genetic Network Programming

AU - Kotaro, Hirasawa

AU - Masafumi, Okubo

AU - Jinglu, Hu

AU - Junichi, Murata

AU - Yuko, Matsuya

PY - 2003/1

Y1 - 2003/1

N2 - Recently, many methods of evolutionary computation such as Genetic Algorithm(GA) and Genetic Programming(GP) have been developed as a basic tool for modeling and optimizing complex systems. GA has the genome of string structure, while the genome in GP is of tree structure. In this paper, a new evolutionary method named Genetic Network Programming(GNP), whose genome has network structure is applied to multiagent sysytems. Hetero Multiagent Sysytems with GNP are studied, where each agent of the same group has its own GNP program in order to build the adaptive agents against changing environments. Specifically, the comparison between Hetero Multiagent Systems and conventional Homo Multiagent Sysytems is carried out in simulations on ants behaviors.

AB - Recently, many methods of evolutionary computation such as Genetic Algorithm(GA) and Genetic Programming(GP) have been developed as a basic tool for modeling and optimizing complex systems. GA has the genome of string structure, while the genome in GP is of tree structure. In this paper, a new evolutionary method named Genetic Network Programming(GNP), whose genome has network structure is applied to multiagent sysytems. Hetero Multiagent Sysytems with GNP are studied, where each agent of the same group has its own GNP program in order to build the adaptive agents against changing environments. Specifically, the comparison between Hetero Multiagent Systems and conventional Homo Multiagent Sysytems is carried out in simulations on ants behaviors.

KW - GA

KW - GNP

KW - GP

KW - artificial life

KW - emergent systems

KW - multi agent system

UR - http://www.scopus.com/inward/record.url?scp=65449117213&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=65449117213&partnerID=8YFLogxK

U2 - 10.1541/ieejeiss.123.544

DO - 10.1541/ieejeiss.123.544

M3 - Article

AN - SCOPUS:65449117213

SN - 0385-4221

VL - 123

SP - 544

EP - 551

JO - IEEJ Transactions on Electronics, Information and Systems

JF - IEEJ Transactions on Electronics, Information and Systems

IS - 3

ER -

Co-evolution of Hetero Multiagent Systems using Genetic Network Programming

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this