Adaptation of neural agent in dynamic environment

Hybrid system of genetic algorithm and neural network

Research output: Chapter in Book/Report/Conference proceedingChapter

7 Citations (Scopus)

Abstract

The purpose of this study is to propose the adaptive agent as hybrid of Genetic Algorithm and Neural Network, and to clarify the effectiveness of the combination of two mechanisms in the dynamic environment. Evolution and learning can be explained as the mechanism of searching a solution in the enormous possibilities at the population level and individual level respectively. Genetic algorithm and neural network are computational models. Genetic algorithm is suitable for global search, and neural network at the local search. Combination of genetic algorithm and neural network seems natural from the biological viewpoint. There are two ways of combination of genetic algorithm and neural network, that is Darwinian and Lamarckian framework. In Lamarckian framework the acquired traits during the lifetime can be passed on to the offspring directly, and in Darwinian framework, these cannot be passed on. We propose `Neural Agent' whose initial weights of their neural networks are determined by their genome data, as a simple model of hybrid system of genetic algorithm and neural network. We examine which framework is better in the dynamic system. The result of our simulation shows Darwinian framework is better than Lamarckian.

Original languageEnglish
Title of host publicationInternational Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES
PublisherIEEE
Pages575-584
Number of pages10
Volume3
Publication statusPublished - 1998
EventProceedings of the 1998 2nd International Conference on knowledge-Based Intelligent Electronic Systems (KES '98) - Adelaide, Aust
Duration: 1998 Apr 211998 Apr 23

Other

OtherProceedings of the 1998 2nd International Conference on knowledge-Based Intelligent Electronic Systems (KES '98)
CityAdelaide, Aust
Period98/4/2198/4/23

Fingerprint

Hybrid systems
Genetic algorithms
Neural networks
Dynamical systems
Genes

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Iba, T., & Takefuji, Y. (1998). Adaptation of neural agent in dynamic environment: Hybrid system of genetic algorithm and neural network. In International Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES (Vol. 3, pp. 575-584). IEEE.

Adaptation of neural agent in dynamic environment : Hybrid system of genetic algorithm and neural network. / Iba, Takashi; Takefuji, Yoshiyasu.

International Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES. Vol. 3 IEEE, 1998. p. 575-584.

Research output: Chapter in Book/Report/Conference proceedingChapter

Iba, T & Takefuji, Y 1998, Adaptation of neural agent in dynamic environment: Hybrid system of genetic algorithm and neural network. in International Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES. vol. 3, IEEE, pp. 575-584, Proceedings of the 1998 2nd International Conference on knowledge-Based Intelligent Electronic Systems (KES '98), Adelaide, Aust, 98/4/21.
Iba T, Takefuji Y. Adaptation of neural agent in dynamic environment: Hybrid system of genetic algorithm and neural network. In International Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES. Vol. 3. IEEE. 1998. p. 575-584
Iba, Takashi ; Takefuji, Yoshiyasu. / Adaptation of neural agent in dynamic environment : Hybrid system of genetic algorithm and neural network. International Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES. Vol. 3 IEEE, 1998. pp. 575-584
@inbook{776898c7a7534602b672c5aef12e33ac,
title = "Adaptation of neural agent in dynamic environment: Hybrid system of genetic algorithm and neural network",
abstract = "The purpose of this study is to propose the adaptive agent as hybrid of Genetic Algorithm and Neural Network, and to clarify the effectiveness of the combination of two mechanisms in the dynamic environment. Evolution and learning can be explained as the mechanism of searching a solution in the enormous possibilities at the population level and individual level respectively. Genetic algorithm and neural network are computational models. Genetic algorithm is suitable for global search, and neural network at the local search. Combination of genetic algorithm and neural network seems natural from the biological viewpoint. There are two ways of combination of genetic algorithm and neural network, that is Darwinian and Lamarckian framework. In Lamarckian framework the acquired traits during the lifetime can be passed on to the offspring directly, and in Darwinian framework, these cannot be passed on. We propose `Neural Agent' whose initial weights of their neural networks are determined by their genome data, as a simple model of hybrid system of genetic algorithm and neural network. We examine which framework is better in the dynamic system. The result of our simulation shows Darwinian framework is better than Lamarckian.",
author = "Takashi Iba and Yoshiyasu Takefuji",
year = "1998",
language = "English",
volume = "3",
pages = "575--584",
booktitle = "International Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES",
publisher = "IEEE",

}

TY - CHAP

T1 - Adaptation of neural agent in dynamic environment

T2 - Hybrid system of genetic algorithm and neural network

AU - Iba, Takashi

AU - Takefuji, Yoshiyasu

PY - 1998

Y1 - 1998

N2 - The purpose of this study is to propose the adaptive agent as hybrid of Genetic Algorithm and Neural Network, and to clarify the effectiveness of the combination of two mechanisms in the dynamic environment. Evolution and learning can be explained as the mechanism of searching a solution in the enormous possibilities at the population level and individual level respectively. Genetic algorithm and neural network are computational models. Genetic algorithm is suitable for global search, and neural network at the local search. Combination of genetic algorithm and neural network seems natural from the biological viewpoint. There are two ways of combination of genetic algorithm and neural network, that is Darwinian and Lamarckian framework. In Lamarckian framework the acquired traits during the lifetime can be passed on to the offspring directly, and in Darwinian framework, these cannot be passed on. We propose `Neural Agent' whose initial weights of their neural networks are determined by their genome data, as a simple model of hybrid system of genetic algorithm and neural network. We examine which framework is better in the dynamic system. The result of our simulation shows Darwinian framework is better than Lamarckian.

AB - The purpose of this study is to propose the adaptive agent as hybrid of Genetic Algorithm and Neural Network, and to clarify the effectiveness of the combination of two mechanisms in the dynamic environment. Evolution and learning can be explained as the mechanism of searching a solution in the enormous possibilities at the population level and individual level respectively. Genetic algorithm and neural network are computational models. Genetic algorithm is suitable for global search, and neural network at the local search. Combination of genetic algorithm and neural network seems natural from the biological viewpoint. There are two ways of combination of genetic algorithm and neural network, that is Darwinian and Lamarckian framework. In Lamarckian framework the acquired traits during the lifetime can be passed on to the offspring directly, and in Darwinian framework, these cannot be passed on. We propose `Neural Agent' whose initial weights of their neural networks are determined by their genome data, as a simple model of hybrid system of genetic algorithm and neural network. We examine which framework is better in the dynamic system. The result of our simulation shows Darwinian framework is better than Lamarckian.

UR - http://www.scopus.com/inward/record.url?scp=0032218851&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032218851&partnerID=8YFLogxK

M3 - Chapter

VL - 3

SP - 575

EP - 584

BT - International Conference on Knowledge-Based Intelligent Electronic Systems, Proceedings, KES

PB - IEEE

ER -