Model-based neural networks

Terry M. Caelli; David Mc G. Squire; Tom P.J. Wild

doi:10.1016/S0893-6080(05)80106-6

Model-based neural networks

Terry M. Caelli, David Mc G. Squire, Tom P.J. Wild

Research output: Contribution to journal › Article › Research › peer-review

27 Citations (Scopus)

Abstract

In this paper we show how neural networks can be formulated in terms of various parameterised connection models which explicitly encode desired properties of the target system. Such a modelling approach to neural networks raises issues about their relationships to other technologies such as Adaptive Filtering and Principal Components Analysis. The benefits of this approach can be a significant decrease in the parameter space, improved generalisation, and a learning procedure which guarantees a priori specified invariance constraints.

Original language	English
Pages (from-to)	613-625
Number of pages	13
Journal	Neural Networks
Volume	6
Issue number	5
DOIs	https://doi.org/10.1016/S0893-6080(05)80106-6
Publication status	Published - 1 Jan 1993
Externally published	Yes

Keywords

Adaptive filters
Classification
Invariance
Number of parameters

Access to Document

10.1016/S0893-6080(05)80106-6

Cite this

@article{6d96b967baf14fab86fdb451441a82bd,

title = "Model-based neural networks",

abstract = "In this paper we show how neural networks can be formulated in terms of various parameterised connection models which explicitly encode desired properties of the target system. Such a modelling approach to neural networks raises issues about their relationships to other technologies such as Adaptive Filtering and Principal Components Analysis. The benefits of this approach can be a significant decrease in the parameter space, improved generalisation, and a learning procedure which guarantees a priori specified invariance constraints.",

keywords = "Adaptive filters, Classification, Invariance, Number of parameters",

author = "Caelli, {Terry M.} and Squire, {David Mc G.} and Wild, {Tom P.J.}",

year = "1993",

month = jan,

day = "1",

doi = "10.1016/S0893-6080(05)80106-6",

language = "English",

volume = "6",

pages = "613--625",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier",

number = "5",

}

TY - JOUR

T1 - Model-based neural networks

AU - Caelli, Terry M.

AU - Squire, David Mc G.

AU - Wild, Tom P.J.

PY - 1993/1/1

Y1 - 1993/1/1

N2 - In this paper we show how neural networks can be formulated in terms of various parameterised connection models which explicitly encode desired properties of the target system. Such a modelling approach to neural networks raises issues about their relationships to other technologies such as Adaptive Filtering and Principal Components Analysis. The benefits of this approach can be a significant decrease in the parameter space, improved generalisation, and a learning procedure which guarantees a priori specified invariance constraints.

AB - In this paper we show how neural networks can be formulated in terms of various parameterised connection models which explicitly encode desired properties of the target system. Such a modelling approach to neural networks raises issues about their relationships to other technologies such as Adaptive Filtering and Principal Components Analysis. The benefits of this approach can be a significant decrease in the parameter space, improved generalisation, and a learning procedure which guarantees a priori specified invariance constraints.

KW - Adaptive filters

KW - Classification

KW - Invariance

KW - Number of parameters

UR - http://www.scopus.com/inward/record.url?scp=0027167949&partnerID=8YFLogxK

U2 - 10.1016/S0893-6080(05)80106-6

DO - 10.1016/S0893-6080(05)80106-6

M3 - Article

AN - SCOPUS:0027167949

VL - 6

SP - 613

EP - 625

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

IS - 5

ER -

Model-based neural networks

Abstract

Keywords

Access to Document

Other files and links

Cite this