Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method

Tomoharu NAGAO; Takeshi AGUI; Hiroshi NAGAHASHI

Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method

Tomoharu NAGAO, Takeshi AGUI, Hiroshi NAGAHASHI

Full Text Views

0

Cite this

Summary :

A genetic method to generate a neural network which has both structure and connection weights adequate for a given task is proposed. A neural network having arbitrary connections is regarded as a virtual living thing which has genes representing its connections among neural units. Effectiveness of the network is estimated from its time sequential input and output signals. Excellent individuals, namely appropriate neural networks, are generated through generation iterations. The basic principle of the method and its applications are described. As an example of evolution from randomly generated networks to feedforward networks, an XOR problem is dealt with, and an action control problem is used for making networks containing feedback and mutual connections. The proposed method is available for designing a neural network whose adequate structure is unknown.

Publication: IEICE TRANSACTIONS on Information Vol.E76-D No.6 pp.689-697

Publication Date: 1993/06/25

Publicized

Online ISSN

DOI

Type of Manuscript: PAPER

Category: Bio-Cybernetics

Cite this

Copy

Tomoharu NAGAO, Takeshi AGUI, Hiroshi NAGAHASHI, "Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method" in IEICE TRANSACTIONS on Information, vol. E76-D, no. 6, pp. 689-697, June 1993, doi: .
Abstract: A genetic method to generate a neural network which has both structure and connection weights adequate for a given task is proposed. A neural network having arbitrary connections is regarded as a virtual living thing which has genes representing its connections among neural units. Effectiveness of the network is estimated from its time sequential input and output signals. Excellent individuals, namely appropriate neural networks, are generated through generation iterations. The basic principle of the method and its applications are described. As an example of evolution from randomly generated networks to feedforward networks, an XOR problem is dealt with, and an action control problem is used for making networks containing feedback and mutual connections. The proposed method is available for designing a neural network whose adequate structure is unknown.
URL: https://global.ieice.org/en_transactions/information/10.1587/e76-d_6_689/_p

Copy

@ARTICLE{e76-d_6_689,
author={Tomoharu NAGAO, Takeshi AGUI, Hiroshi NAGAHASHI, },
journal={IEICE TRANSACTIONS on Information},
title={Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method},
year={1993},
volume={E76-D},
number={6},
pages={689-697},
abstract={A genetic method to generate a neural network which has both structure and connection weights adequate for a given task is proposed. A neural network having arbitrary connections is regarded as a virtual living thing which has genes representing its connections among neural units. Effectiveness of the network is estimated from its time sequential input and output signals. Excellent individuals, namely appropriate neural networks, are generated through generation iterations. The basic principle of the method and its applications are described. As an example of evolution from randomly generated networks to feedforward networks, an XOR problem is dealt with, and an action control problem is used for making networks containing feedback and mutual connections. The proposed method is available for designing a neural network whose adequate structure is unknown.},
keywords={},
doi={},
ISSN={},
month={June},}

Copy

TY - JOUR
TI - Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method
T2 - IEICE TRANSACTIONS on Information
SP - 689
EP - 697
AU - Tomoharu NAGAO
AU - Takeshi AGUI
AU - Hiroshi NAGAHASHI
PY - 1993
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E76-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 1993
AB - A genetic method to generate a neural network which has both structure and connection weights adequate for a given task is proposed. A neural network having arbitrary connections is regarded as a virtual living thing which has genes representing its connections among neural units. Effectiveness of the network is estimated from its time sequential input and output signals. Excellent individuals, namely appropriate neural networks, are generated through generation iterations. The basic principle of the method and its applications are described. As an example of evolution from randomly generated networks to feedforward networks, an XOR problem is dealt with, and an action control problem is used for making networks containing feedback and mutual connections. The proposed method is available for designing a neural network whose adequate structure is unknown.
ER -