Tuning of the structure and parameters of neural network using an improved genetic algorithm

H. K. Lam; S. H. Ling; Hung Fat Frank Leung; P. K S Tam

Tuning of the structure and parameters of neural network using an improved genetic algorithm

H. K. Lam, S. H. Ling, Hung Fat Frank Leung, P. K S Tam

The Hong Kong Polytechnic University

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

49 Citations (Scopus)

Abstract

This paper presents the tuning of the structure and parameters of a neural network using an improved genetic algorithm (GA). The improved GA is implemented by floating-point number. The processing time of the improved GA is faster than that of the GA implemented by binary number as coding and decoding are not necessary. By introducing new genetic operators to the improved GA, it will also be shown that the improved GA performs better than the traditional GA based on some benchmark test functions. A neural network with switches introduced to links is proposed. By doing this, the proposed neural network can learn both the input-output relationships of an application and the network structure. Using the improved GA, the structure and the parameters of the neural network can be tuned. An application example on sunspot forecasting is given to show the merits of the improved GA and the proposed neural network.

Original language	English
Title of host publication	IECON Proceedings (Industrial Electronics Conference)
Pages	25-30
Number of pages	6
Publication status	Published - 1 Dec 2001
Event	27th Annual Conference of the IEEE Industrial Electronics Society IECON'2001 - Denver, CO, United States Duration: 29 Nov 2001 → 2 Dec 2001

Conference

Conference	27th Annual Conference of the IEEE Industrial Electronics Society IECON'2001
Country/Territory	United States
City	Denver, CO
Period	29/11/01 → 2/12/01

ASJC Scopus subject areas

Electrical and Electronic Engineering

More information

http://hdl.handle.net/10397/1371

Cite this

@inproceedings{c82eb835ea4343679681b63a0f70b3e2,

title = "Tuning of the structure and parameters of neural network using an improved genetic algorithm",

abstract = "This paper presents the tuning of the structure and parameters of a neural network using an improved genetic algorithm (GA). The improved GA is implemented by floating-point number. The processing time of the improved GA is faster than that of the GA implemented by binary number as coding and decoding are not necessary. By introducing new genetic operators to the improved GA, it will also be shown that the improved GA performs better than the traditional GA based on some benchmark test functions. A neural network with switches introduced to links is proposed. By doing this, the proposed neural network can learn both the input-output relationships of an application and the network structure. Using the improved GA, the structure and the parameters of the neural network can be tuned. An application example on sunspot forecasting is given to show the merits of the improved GA and the proposed neural network.",

author = "Lam, {H. K.} and Ling, {S. H.} and Leung, {Hung Fat Frank} and Tam, {P. K S}",

year = "2001",

month = dec,

day = "1",

language = "English",

pages = "25--30",

booktitle = "IECON Proceedings (Industrial Electronics Conference)",

note = "27th Annual Conference of the IEEE Industrial Electronics Society IECON'2001 ; Conference date: 29-11-2001 Through 02-12-2001",

}

TY - GEN

T1 - Tuning of the structure and parameters of neural network using an improved genetic algorithm

AU - Lam, H. K.

AU - Ling, S. H.

AU - Leung, Hung Fat Frank

AU - Tam, P. K S

PY - 2001/12/1

Y1 - 2001/12/1

N2 - This paper presents the tuning of the structure and parameters of a neural network using an improved genetic algorithm (GA). The improved GA is implemented by floating-point number. The processing time of the improved GA is faster than that of the GA implemented by binary number as coding and decoding are not necessary. By introducing new genetic operators to the improved GA, it will also be shown that the improved GA performs better than the traditional GA based on some benchmark test functions. A neural network with switches introduced to links is proposed. By doing this, the proposed neural network can learn both the input-output relationships of an application and the network structure. Using the improved GA, the structure and the parameters of the neural network can be tuned. An application example on sunspot forecasting is given to show the merits of the improved GA and the proposed neural network.

AB - This paper presents the tuning of the structure and parameters of a neural network using an improved genetic algorithm (GA). The improved GA is implemented by floating-point number. The processing time of the improved GA is faster than that of the GA implemented by binary number as coding and decoding are not necessary. By introducing new genetic operators to the improved GA, it will also be shown that the improved GA performs better than the traditional GA based on some benchmark test functions. A neural network with switches introduced to links is proposed. By doing this, the proposed neural network can learn both the input-output relationships of an application and the network structure. Using the improved GA, the structure and the parameters of the neural network can be tuned. An application example on sunspot forecasting is given to show the merits of the improved GA and the proposed neural network.

UR - http://www.scopus.com/inward/record.url?scp=0035690148&partnerID=8YFLogxK

M3 - Conference article published in proceeding or book

SP - 25

EP - 30

BT - IECON Proceedings (Industrial Electronics Conference)

T2 - 27th Annual Conference of the IEEE Industrial Electronics Society IECON'2001

Y2 - 29 November 2001 through 2 December 2001

ER -

Tuning of the structure and parameters of neural network using an improved genetic algorithm

Abstract

Conference

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this