Evolving neural networks using swarm intelligence for binmap classification

Emilio Miguelañez, A. M S Zalzala, Paul Tabor

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A novel automatic defect classification system is introduced for electrical test analysis of semiconductor wafer using evolutionary algorithm techniques to construct Radial Basis Function Neural Networks (RBF NNs) as a classifier. The parameters of a RBF NN (number of neurons, and their respective centers and radii) are often determined by hand or based on methods highly dependent on initial values. In this work, Particle Swarm Optimization algorithm is implemented to build a RBF NN that solves this specific problem. As a primary input source to the network, the system employs electrical binmaps obtained from the test stage of the manufacturing process. To accomplish this task, a filtering algorithm is also implemented able to discard those wafermaps without pattern. The performance of the reported approach shows an outstanding e-bitmap classification rate. To evaluate the performance of the main algorithm, the system is tested also on the Australian credit card data set and the error rate obtained is comparable with the best algorithms found in the literature.

Original languageEnglish
Title of host publicationProceedings of the 2004 Congress on Evolutionary Computation, CEC2004
Pages978-985
Number of pages8
Volume1
Publication statusPublished - 2004
EventProceedings of the 2004 Congress on Evolutionary Computation, CEC2004 - Portland, OR, United States
Duration: 19 Jun 200423 Jun 2004

Conference

ConferenceProceedings of the 2004 Congress on Evolutionary Computation, CEC2004
CountryUnited States
CityPortland, OR
Period19/06/0423/06/04

Fingerprint

Neural networks
Evolutionary algorithms
Particle swarm optimization (PSO)
Neurons
Classifiers
Semiconductor materials
Defects
Swarm intelligence

Cite this

Miguelañez, E., Zalzala, A. M. S., & Tabor, P. (2004). Evolving neural networks using swarm intelligence for binmap classification. In Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004 (Vol. 1, pp. 978-985)
Miguelañez, Emilio ; Zalzala, A. M S ; Tabor, Paul. / Evolving neural networks using swarm intelligence for binmap classification. Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. Vol. 1 2004. pp. 978-985
@inproceedings{31ee30a7daf24b7483a6d16021061075,
title = "Evolving neural networks using swarm intelligence for binmap classification",
abstract = "A novel automatic defect classification system is introduced for electrical test analysis of semiconductor wafer using evolutionary algorithm techniques to construct Radial Basis Function Neural Networks (RBF NNs) as a classifier. The parameters of a RBF NN (number of neurons, and their respective centers and radii) are often determined by hand or based on methods highly dependent on initial values. In this work, Particle Swarm Optimization algorithm is implemented to build a RBF NN that solves this specific problem. As a primary input source to the network, the system employs electrical binmaps obtained from the test stage of the manufacturing process. To accomplish this task, a filtering algorithm is also implemented able to discard those wafermaps without pattern. The performance of the reported approach shows an outstanding e-bitmap classification rate. To evaluate the performance of the main algorithm, the system is tested also on the Australian credit card data set and the error rate obtained is comparable with the best algorithms found in the literature.",
author = "Emilio Miguela{\~n}ez and Zalzala, {A. M S} and Paul Tabor",
year = "2004",
language = "English",
isbn = "0780385152",
volume = "1",
pages = "978--985",
booktitle = "Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004",

}

Miguelañez, E, Zalzala, AMS & Tabor, P 2004, Evolving neural networks using swarm intelligence for binmap classification. in Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. vol. 1, pp. 978-985, Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004, Portland, OR, United States, 19/06/04.

Evolving neural networks using swarm intelligence for binmap classification. / Miguelañez, Emilio; Zalzala, A. M S; Tabor, Paul.

Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. Vol. 1 2004. p. 978-985.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Evolving neural networks using swarm intelligence for binmap classification

AU - Miguelañez, Emilio

AU - Zalzala, A. M S

AU - Tabor, Paul

PY - 2004

Y1 - 2004

N2 - A novel automatic defect classification system is introduced for electrical test analysis of semiconductor wafer using evolutionary algorithm techniques to construct Radial Basis Function Neural Networks (RBF NNs) as a classifier. The parameters of a RBF NN (number of neurons, and their respective centers and radii) are often determined by hand or based on methods highly dependent on initial values. In this work, Particle Swarm Optimization algorithm is implemented to build a RBF NN that solves this specific problem. As a primary input source to the network, the system employs electrical binmaps obtained from the test stage of the manufacturing process. To accomplish this task, a filtering algorithm is also implemented able to discard those wafermaps without pattern. The performance of the reported approach shows an outstanding e-bitmap classification rate. To evaluate the performance of the main algorithm, the system is tested also on the Australian credit card data set and the error rate obtained is comparable with the best algorithms found in the literature.

AB - A novel automatic defect classification system is introduced for electrical test analysis of semiconductor wafer using evolutionary algorithm techniques to construct Radial Basis Function Neural Networks (RBF NNs) as a classifier. The parameters of a RBF NN (number of neurons, and their respective centers and radii) are often determined by hand or based on methods highly dependent on initial values. In this work, Particle Swarm Optimization algorithm is implemented to build a RBF NN that solves this specific problem. As a primary input source to the network, the system employs electrical binmaps obtained from the test stage of the manufacturing process. To accomplish this task, a filtering algorithm is also implemented able to discard those wafermaps without pattern. The performance of the reported approach shows an outstanding e-bitmap classification rate. To evaluate the performance of the main algorithm, the system is tested also on the Australian credit card data set and the error rate obtained is comparable with the best algorithms found in the literature.

UR - http://www.scopus.com/inward/record.url?scp=4344694785&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0780385152

SN - 9780780385153

VL - 1

SP - 978

EP - 985

BT - Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004

ER -

Miguelañez E, Zalzala AMS, Tabor P. Evolving neural networks using swarm intelligence for binmap classification. In Proceedings of the 2004 Congress on Evolutionary Computation, CEC2004. Vol. 1. 2004. p. 978-985