A pipelined speculative SIMD architecture for SOM ANN

O. Hammami; D. Suzuki

doi:10.1109/ICNN.1997.616160

A pipelined speculative SIMD architecture for SOM ANN

O. Hammami, D. Suzuki

Univ of Aizu

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

Abstract

Hardware implementations of the self-organising map (SOM) ANN have often favored a standard SIMD type of architecture. Although the SOM have some features to support this choice, the standard SIMD model does not sufficiently exploit temporal parallelism and does not exploit the idle time of the idle neurons outside of the winning neuron neighborhood during weights updating. Some architectural proposals are made and analyzed in this context based on digital synchronous design techniques.

Original language	English
Title of host publication	1997 IEEE International Conference on Neural Networks, ICNN 1997
Pages	985-990
Number of pages	6
DOIs	https://doi.org/10.1109/ICNN.1997.616160
Publication status	Published - 1 Dec 1997
Externally published	Yes
Event	1997 IEEE International Conference on Neural Networks, ICNN 1997 - Houston, TX, United States Duration: 9 Jun 1997 → 12 Jun 1997

Publication series

Name	IEEE International Conference on Neural Networks - Conference Proceedings
Volume	2
ISSN (Print)	1098-7576

Conference

Conference	1997 IEEE International Conference on Neural Networks, ICNN 1997
Country/Territory	United States
City	Houston, TX
Period	9/06/97 → 12/06/97

Access to Document

10.1109/ICNN.1997.616160

Cite this

@inproceedings{4bace08f7e0c4ed7b85671e0cf06abde,

title = "A pipelined speculative SIMD architecture for SOM ANN",

abstract = "Hardware implementations of the self-organising map (SOM) ANN have often favored a standard SIMD type of architecture. Although the SOM have some features to support this choice, the standard SIMD model does not sufficiently exploit temporal parallelism and does not exploit the idle time of the idle neurons outside of the winning neuron neighborhood during weights updating. Some architectural proposals are made and analyzed in this context based on digital synchronous design techniques.",

author = "O. Hammami and D. Suzuki",

year = "1997",

month = dec,

day = "1",

doi = "10.1109/ICNN.1997.616160",

language = "English",

isbn = "0780341228",

series = "IEEE International Conference on Neural Networks - Conference Proceedings",

pages = "985--990",

booktitle = "1997 IEEE International Conference on Neural Networks, ICNN 1997",

note = "1997 IEEE International Conference on Neural Networks, ICNN 1997 ; Conference date: 09-06-1997 Through 12-06-1997",

}

Hammami, O & Suzuki, D 1997, A pipelined speculative SIMD architecture for SOM ANN. in 1997 IEEE International Conference on Neural Networks, ICNN 1997., 616160, IEEE International Conference on Neural Networks - Conference Proceedings, vol. 2, pp. 985-990, 1997 IEEE International Conference on Neural Networks, ICNN 1997, Houston, TX, United States, 9/06/97. https://doi.org/10.1109/ICNN.1997.616160

TY - GEN

T1 - A pipelined speculative SIMD architecture for SOM ANN

AU - Hammami, O.

AU - Suzuki, D.

PY - 1997/12/1

Y1 - 1997/12/1

N2 - Hardware implementations of the self-organising map (SOM) ANN have often favored a standard SIMD type of architecture. Although the SOM have some features to support this choice, the standard SIMD model does not sufficiently exploit temporal parallelism and does not exploit the idle time of the idle neurons outside of the winning neuron neighborhood during weights updating. Some architectural proposals are made and analyzed in this context based on digital synchronous design techniques.

AB - Hardware implementations of the self-organising map (SOM) ANN have often favored a standard SIMD type of architecture. Although the SOM have some features to support this choice, the standard SIMD model does not sufficiently exploit temporal parallelism and does not exploit the idle time of the idle neurons outside of the winning neuron neighborhood during weights updating. Some architectural proposals are made and analyzed in this context based on digital synchronous design techniques.

U2 - 10.1109/ICNN.1997.616160

DO - 10.1109/ICNN.1997.616160

M3 - Conference contribution

AN - SCOPUS:0030651572

SN - 0780341228

SN - 9780780341227

T3 - IEEE International Conference on Neural Networks - Conference Proceedings

SP - 985

EP - 990

BT - 1997 IEEE International Conference on Neural Networks, ICNN 1997

T2 - 1997 IEEE International Conference on Neural Networks, ICNN 1997

Y2 - 9 June 1997 through 12 June 1997

ER -

A pipelined speculative SIMD architecture for SOM ANN

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this