An evolutionary genetic neural networks for problems without prior knowledge

Hyoung Uk Ha; Jong-Kook Kim

doi:10.1109/ICNC.2014.6975800

An evolutionary genetic neural networks for problems without prior knowledge

Hyoung Uk Ha, Jong-Kook Kim

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

Many problems are now being solved using a version of a neural network (NN). These NN are usually constructed using genetic neural networks (GNNs) for optimizing variables in the NN using a fixed structure or neural evolution (NE) to optimize the structure of the NN using fixed values for the variables in the NN. Thus, previous methods need experienced knowledge of the problem such that either the structure or variables are known to construct a meaningful NN. This paper presents a method called leap evolution adopted neural network (LEANN) that optimizes the NN without prior knowledge such as the values of the variables and the structure of the NN for a given problem. Our method in this paper finds an optimal structure and variables of the NN successfully for the XOR gate problem.

Original language	English
Title of host publication	2014 10th International Conference on Natural Computation, ICNC 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-6
Number of pages	6
ISBN (Print)	9781479951505
DOIs	https://doi.org/10.1109/ICNC.2014.6975800
Publication status	Published - 2014
Event	2014 10th International Conference on Natural Computation, ICNC 2014 - Xiamen, China Duration: 2014 Aug 19 → 2014 Aug 21

Other

Other	2014 10th International Conference on Natural Computation, ICNC 2014
Country/Territory	China
City	Xiamen
Period	14/8/19 → 14/8/21

Keywords

Bio-inspired algorithm
Evolutionary algorithm
Genetic algorithm
Genetic neural network
Multilayer perceptron
Neuro-evolution

ASJC Scopus subject areas

Artificial Intelligence
Computational Theory and Mathematics
Computer Networks and Communications
Computer Science Applications
Computer Vision and Pattern Recognition
Signal Processing
Biomedical Engineering

Access to Document

10.1109/ICNC.2014.6975800

Cite this

Ha, HU & Kim, J-K 2014, An evolutionary genetic neural networks for problems without prior knowledge. in 2014 10th International Conference on Natural Computation, ICNC 2014., 6975800, Institute of Electrical and Electronics Engineers Inc., pp. 1-6, 2014 10th International Conference on Natural Computation, ICNC 2014, Xiamen, China, 14/8/19. https://doi.org/10.1109/ICNC.2014.6975800

@inproceedings{3c73415ef36f4c43815a7a1a189acca4,

title = "An evolutionary genetic neural networks for problems without prior knowledge",

abstract = "Many problems are now being solved using a version of a neural network (NN). These NN are usually constructed using genetic neural networks (GNNs) for optimizing variables in the NN using a fixed structure or neural evolution (NE) to optimize the structure of the NN using fixed values for the variables in the NN. Thus, previous methods need experienced knowledge of the problem such that either the structure or variables are known to construct a meaningful NN. This paper presents a method called leap evolution adopted neural network (LEANN) that optimizes the NN without prior knowledge such as the values of the variables and the structure of the NN for a given problem. Our method in this paper finds an optimal structure and variables of the NN successfully for the XOR gate problem.",

keywords = "Bio-inspired algorithm, Evolutionary algorithm, Genetic algorithm, Genetic neural network, Multilayer perceptron, Neuro-evolution",

author = "Ha, {Hyoung Uk} and Jong-Kook Kim",

year = "2014",

doi = "10.1109/ICNC.2014.6975800",

language = "English",

isbn = "9781479951505",

pages = "1--6",

booktitle = "2014 10th International Conference on Natural Computation, ICNC 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "2014 10th International Conference on Natural Computation, ICNC 2014 ; Conference date: 19-08-2014 Through 21-08-2014",

}

TY - GEN

T1 - An evolutionary genetic neural networks for problems without prior knowledge

AU - Ha, Hyoung Uk

AU - Kim, Jong-Kook

PY - 2014

Y1 - 2014

N2 - Many problems are now being solved using a version of a neural network (NN). These NN are usually constructed using genetic neural networks (GNNs) for optimizing variables in the NN using a fixed structure or neural evolution (NE) to optimize the structure of the NN using fixed values for the variables in the NN. Thus, previous methods need experienced knowledge of the problem such that either the structure or variables are known to construct a meaningful NN. This paper presents a method called leap evolution adopted neural network (LEANN) that optimizes the NN without prior knowledge such as the values of the variables and the structure of the NN for a given problem. Our method in this paper finds an optimal structure and variables of the NN successfully for the XOR gate problem.

AB - Many problems are now being solved using a version of a neural network (NN). These NN are usually constructed using genetic neural networks (GNNs) for optimizing variables in the NN using a fixed structure or neural evolution (NE) to optimize the structure of the NN using fixed values for the variables in the NN. Thus, previous methods need experienced knowledge of the problem such that either the structure or variables are known to construct a meaningful NN. This paper presents a method called leap evolution adopted neural network (LEANN) that optimizes the NN without prior knowledge such as the values of the variables and the structure of the NN for a given problem. Our method in this paper finds an optimal structure and variables of the NN successfully for the XOR gate problem.

KW - Bio-inspired algorithm

KW - Evolutionary algorithm

KW - Genetic algorithm

KW - Genetic neural network

KW - Multilayer perceptron

KW - Neuro-evolution

UR - http://www.scopus.com/inward/record.url?scp=84926677847&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926677847&partnerID=8YFLogxK

U2 - 10.1109/ICNC.2014.6975800

DO - 10.1109/ICNC.2014.6975800

M3 - Conference contribution

AN - SCOPUS:84926677847

SN - 9781479951505

SP - 1

EP - 6

BT - 2014 10th International Conference on Natural Computation, ICNC 2014

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 10th International Conference on Natural Computation, ICNC 2014

Y2 - 19 August 2014 through 21 August 2014

ER -

An evolutionary genetic neural networks for problems without prior knowledge

Abstract

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this