Fault tolerance and redundancy of a neural net for the classification of acoustic data
Fault tolerance and redundancy of a neural net for the classification of acoustic data
An investigation is made of the relation between the fault tolerance of a multilayer perceptron (MLP) and its redundancy as determined by the number of hidden-layer neurons (x). Damage was introduced by cutting connections. The application studied is the classification of coins according to their acoustic emissions after striking a hard object. Several MLPs were trained by backpropagation to discriminate acoustic emission data from 6 classes of coin. The nets had 259 input nodes, 6 output nodes, and x varying between 5 and 25. In addition, one single-layer network (x=0) was trained. Results show that the single-layer perceptron (SLP)-although able to classify the data with 100% accuracy under fault-free conditions-was far less damage-resistant than any of the MLPs.
1061-1064
Emmerson, M. D.
98374280-664f-4efd-91ee-817a8c8f03a0
Damper, R. I.
6e0e7fdc-57ec-44d4-bc0f-029d17ba441d
Hey, A. J. G.
1e410ccc-7356-4694-89eb-f84410675416
Upstill, C.
43e54278-486b-40e4-9ca1-c63aa8afae19
1991
Emmerson, M. D.
98374280-664f-4efd-91ee-817a8c8f03a0
Damper, R. I.
6e0e7fdc-57ec-44d4-bc0f-029d17ba441d
Hey, A. J. G.
1e410ccc-7356-4694-89eb-f84410675416
Upstill, C.
43e54278-486b-40e4-9ca1-c63aa8afae19
Emmerson, M. D., Damper, R. I., Hey, A. J. G. and Upstill, C.
(1991)
Fault tolerance and redundancy of a neural net for the classification of acoustic data.
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP '91), , Toronto, Canada.
14 - 17 Apr 1991.
.
(doi:10.1109/ICASSP.1991.150529).
Record type:
Conference or Workshop Item
(Paper)
Abstract
An investigation is made of the relation between the fault tolerance of a multilayer perceptron (MLP) and its redundancy as determined by the number of hidden-layer neurons (x). Damage was introduced by cutting connections. The application studied is the classification of coins according to their acoustic emissions after striking a hard object. Several MLPs were trained by backpropagation to discriminate acoustic emission data from 6 classes of coin. The nets had 259 input nodes, 6 output nodes, and x varying between 5 and 25. In addition, one single-layer network (x=0) was trained. Results show that the single-layer perceptron (SLP)-although able to classify the data with 100% accuracy under fault-free conditions-was far less damage-resistant than any of the MLPs.
This record has no associated files available for download.
More information
Published date: 1991
Additional Information:
Published in Lecture Notes in Computer Science In the Proceedings of 1991 IEEE ICASSP Conference.
Venue - Dates:
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP '91), , Toronto, Canada, 1991-04-14 - 1991-04-17
Organisations:
Electronics & Computer Science, IT Innovation, Southampton Wireless Group
Identifiers
Local EPrints ID: 250276
URI: http://eprints.soton.ac.uk/id/eprint/250276
PURE UUID: a933d996-ea58-4dbd-adb2-25632a9038b4
Catalogue record
Date deposited: 27 Jun 2003
Last modified: 14 Mar 2024 04:51
Export record
Altmetrics
Contributors
Author:
M. D. Emmerson
Author:
R. I. Damper
Author:
A. J. G. Hey
Author:
C. Upstill
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics