A Statistical Perspective of Neural Networks for Process Modeling and Control

S. Joe QIN

doi:10.1109/isic.1993.397629

A Statistical Perspective of Neural Networks for Process Modeling and Control

^*Corresponding author for this work

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Referred Conference Paper › peer-review

15 Citations (Scopus)

Abstract

Multilayer neural networks have been successfully applied to industrial process modeling and control. In this paper, the prediction variance of neural networks from gradient based learning is analyzed in the presence of correlated process inputs. Several biased regression approaches, including ridge regression, principal component analysis, and partial least squares, are integrated with neural net training to reduce the prediction variance. Examples are given to illustrate the improvement of the integrated approaches.

Original language	English
Title of host publication	Proceedings of the 8th IEEE International Symposium on Intelligent Control, Chicago, Illinois, USA - August 1993
Publisher	Institute of Electrical and Electronics Engineers
Pages	599-604
Number of pages	6
ISBN (Print)	0780312066
DOIs	https://doi.org/10.1109/isic.1993.397629
Publication status	Published - Aug 1993
Externally published	Yes
Event	8th IEEE International Symposium on Intelligent Control - Chicago, United States Duration: 25 Aug 1993 → 27 Aug 1993

Conference

Conference	8th IEEE International Symposium on Intelligent Control
Country/Territory	United States
City	Chicago
Period	25/08/93 → 27/08/93

Access to Document

10.1109/isic.1993.397629

Cite this

@inproceedings{2568b08bc2fe486784d9f8104fac5e41,

title = "A Statistical Perspective of Neural Networks for Process Modeling and Control",

abstract = "Multilayer neural networks have been successfully applied to industrial process modeling and control. In this paper, the prediction variance of neural networks from gradient based learning is analyzed in the presence of correlated process inputs. Several biased regression approaches, including ridge regression, principal component analysis, and partial least squares, are integrated with neural net training to reduce the prediction variance. Examples are given to illustrate the improvement of the integrated approaches.",

author = "QIN, {S. Joe}",

year = "1993",

month = aug,

doi = "10.1109/isic.1993.397629",

language = "English",

isbn = "0780312066",

pages = "599--604",

booktitle = "Proceedings of the 8th IEEE International Symposium on Intelligent Control, Chicago, Illinois, USA - August 1993",

publisher = "Institute of Electrical and Electronics Engineers",

note = "8th IEEE International Symposium on Intelligent Control ; Conference date: 25-08-1993 Through 27-08-1993",

}

QIN, SJ 1993, A Statistical Perspective of Neural Networks for Process Modeling and Control. in Proceedings of the 8th IEEE International Symposium on Intelligent Control, Chicago, Illinois, USA - August 1993. Institute of Electrical and Electronics Engineers, pp. 599-604, 8th IEEE International Symposium on Intelligent Control, Chicago, Illinois, United States, 25/08/93. https://doi.org/10.1109/isic.1993.397629

A Statistical Perspective of Neural Networks for Process Modeling and Control. / QIN, S. Joe.
Proceedings of the 8th IEEE International Symposium on Intelligent Control, Chicago, Illinois, USA - August 1993. Institute of Electrical and Electronics Engineers, 1993. p. 599-604.

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Referred Conference Paper › peer-review

TY - GEN

T1 - A Statistical Perspective of Neural Networks for Process Modeling and Control

AU - QIN, S. Joe

PY - 1993/8

Y1 - 1993/8

N2 - Multilayer neural networks have been successfully applied to industrial process modeling and control. In this paper, the prediction variance of neural networks from gradient based learning is analyzed in the presence of correlated process inputs. Several biased regression approaches, including ridge regression, principal component analysis, and partial least squares, are integrated with neural net training to reduce the prediction variance. Examples are given to illustrate the improvement of the integrated approaches.

AB - Multilayer neural networks have been successfully applied to industrial process modeling and control. In this paper, the prediction variance of neural networks from gradient based learning is analyzed in the presence of correlated process inputs. Several biased regression approaches, including ridge regression, principal component analysis, and partial least squares, are integrated with neural net training to reduce the prediction variance. Examples are given to illustrate the improvement of the integrated approaches.

UR - http://www.scopus.com/inward/record.url?scp=0027803588&partnerID=8YFLogxK

U2 - 10.1109/isic.1993.397629

DO - 10.1109/isic.1993.397629

M3 - Conference paper (refereed)

SN - 0780312066

SP - 599

EP - 604

BT - Proceedings of the 8th IEEE International Symposium on Intelligent Control, Chicago, Illinois, USA - August 1993

PB - Institute of Electrical and Electronics Engineers

T2 - 8th IEEE International Symposium on Intelligent Control

Y2 - 25 August 1993 through 27 August 1993

ER -

A Statistical Perspective of Neural Networks for Process Modeling and Control

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this