An introduction to modular map systems

N. Lightowler, C. T. Spracklen, A. R. Allen

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

Abstract

We present an overview of our design for a fully digital hardware implementation of the Self Organising Map (SOM) (T. Kohonen, 1982). Our approach has resulted in a modular system (Modular Maps) which utilises fine grain parallelism with each neuron being a separate entity implemented as a small RISC processor. The essence of the SOM has been maintained by this design, although minor modifications have been made to the original algorithm to facilitate implementation. Modules can be used as either stand alone systems or combined to enable large networks to be created and large input vectors to be catered for. A simulator system was developed to facilitate investigation into the high level behaviour of Modular Map systems and, as Modular Maps are computationally intensive and parallel in nature, it was implemented on a parallel computer system. A series of simulations was carried out using encoded images of human faces where it was found that the classification accuracy of a Modular Map system offered an improvement over that of the traditional SOM
Original languageEnglish
Title of host publicationNeural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133
PublisherIEE
Pages3/1-3/4
Number of pages4
Volume1997/133
Edition133
DOIs
Publication statusPublished - 1997

Publication series

NameColloqium Digest
PublisherIEE
Number133
Volume1997
ISSN (Print)0963-3308

Fingerprint

Self organizing maps
Reduced instruction set computing
Neurons
Computer systems
Simulators
Hardware

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Lightowler, N., Spracklen, C. T., & Allen, A. R. (1997). An introduction to modular map systems. In Neural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133 (133 ed., Vol. 1997/133, pp. 3/1-3/4). (Colloqium Digest; Vol. 1997, No. 133). IEE . https://doi.org/10.1049/ic:19970732

An introduction to modular map systems. / Lightowler, N.; Spracklen, C. T.; Allen, A. R.

Neural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133. Vol. 1997/133 133. ed. IEE , 1997. p. 3/1-3/4 (Colloqium Digest; Vol. 1997, No. 133).

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

Lightowler, N, Spracklen, CT & Allen, AR 1997, An introduction to modular map systems. in Neural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133. 133 edn, vol. 1997/133, Colloqium Digest, no. 133, vol. 1997, IEE , pp. 3/1-3/4. https://doi.org/10.1049/ic:19970732
Lightowler N, Spracklen CT, Allen AR. An introduction to modular map systems. In Neural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133. 133 ed. Vol. 1997/133. IEE . 1997. p. 3/1-3/4. (Colloqium Digest; 133). https://doi.org/10.1049/ic:19970732
Lightowler, N. ; Spracklen, C. T. ; Allen, A. R. / An introduction to modular map systems. Neural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133. Vol. 1997/133 133. ed. IEE , 1997. pp. 3/1-3/4 (Colloqium Digest; 133).
@inproceedings{3df4c42b34e24415912e7de14cee33a1,
title = "An introduction to modular map systems",
abstract = "We present an overview of our design for a fully digital hardware implementation of the Self Organising Map (SOM) (T. Kohonen, 1982). Our approach has resulted in a modular system (Modular Maps) which utilises fine grain parallelism with each neuron being a separate entity implemented as a small RISC processor. The essence of the SOM has been maintained by this design, although minor modifications have been made to the original algorithm to facilitate implementation. Modules can be used as either stand alone systems or combined to enable large networks to be created and large input vectors to be catered for. A simulator system was developed to facilitate investigation into the high level behaviour of Modular Map systems and, as Modular Maps are computationally intensive and parallel in nature, it was implemented on a parallel computer system. A series of simulations was carried out using encoded images of human faces where it was found that the classification accuracy of a Modular Map system offered an improvement over that of the traditional SOM",
author = "N. Lightowler and Spracklen, {C. T.} and Allen, {A. R.}",
year = "1997",
doi = "10.1049/ic:19970732",
language = "English",
volume = "1997/133",
series = "Colloqium Digest",
publisher = "IEE",
number = "133",
pages = "3/1--3/4",
booktitle = "Neural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133",
edition = "133",

}

TY - GEN

T1 - An introduction to modular map systems

AU - Lightowler, N.

AU - Spracklen, C. T.

AU - Allen, A. R.

PY - 1997

Y1 - 1997

N2 - We present an overview of our design for a fully digital hardware implementation of the Self Organising Map (SOM) (T. Kohonen, 1982). Our approach has resulted in a modular system (Modular Maps) which utilises fine grain parallelism with each neuron being a separate entity implemented as a small RISC processor. The essence of the SOM has been maintained by this design, although minor modifications have been made to the original algorithm to facilitate implementation. Modules can be used as either stand alone systems or combined to enable large networks to be created and large input vectors to be catered for. A simulator system was developed to facilitate investigation into the high level behaviour of Modular Map systems and, as Modular Maps are computationally intensive and parallel in nature, it was implemented on a parallel computer system. A series of simulations was carried out using encoded images of human faces where it was found that the classification accuracy of a Modular Map system offered an improvement over that of the traditional SOM

AB - We present an overview of our design for a fully digital hardware implementation of the Self Organising Map (SOM) (T. Kohonen, 1982). Our approach has resulted in a modular system (Modular Maps) which utilises fine grain parallelism with each neuron being a separate entity implemented as a small RISC processor. The essence of the SOM has been maintained by this design, although minor modifications have been made to the original algorithm to facilitate implementation. Modules can be used as either stand alone systems or combined to enable large networks to be created and large input vectors to be catered for. A simulator system was developed to facilitate investigation into the high level behaviour of Modular Map systems and, as Modular Maps are computationally intensive and parallel in nature, it was implemented on a parallel computer system. A series of simulations was carried out using encoded images of human faces where it was found that the classification accuracy of a Modular Map system offered an improvement over that of the traditional SOM

UR - http://www.scopus.com/inward/record.url?scp=6144236523&partnerID=8YFLogxK

U2 - 10.1049/ic:19970732

DO - 10.1049/ic:19970732

M3 - Conference contribution

VL - 1997/133

T3 - Colloqium Digest

SP - 3/1-3/4

BT - Neural and Fuzzy Systems: Design, Hardware and Applications, Colloquium Digest 1997/133

PB - IEE

ER -

Access to Document