Fault Tolerant Training for Optimal Interpolative Nets

Daniel J. Simon; Hossny El-Sherief

doi:10.1109/72.471356

Fault Tolerant Training for Optimal Interpolative Nets

Daniel J. Simon
, Hossny El-Sherief

TRW System Integration Group

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

Abstract

The optimal interpolative (OI) classification network is extended to include fault tolerance and make the network more robust to the loss of a neuron. The OI net has the characteristic that the training data are fit with no more neurons than necessary. Fault tolerance further reduces the number of neurons generated during the learning procedure while maintaining the generalization capabilities of the network. The learning algorithm for the fault-tolerant OI net is presented in a recursive formal, allowing for relatively short training times. A simulated fault-tolerant OI net is tested on a navigation satellite selection problem

Original language	American English
Journal	IEEE Transactions on Neural Networks
Volume	6
DOIs	https://doi.org/10.1109/72.471356
State	Published - Nov 1 1995

Keywords

Classification network
Fault-tolerant training
Generalization
Learning procedure
Navigation satellite selection
Neural nets
Optimal interpolative nets

Disciplines

Electrical and Computer Engineering

Access to Document

10.1109/72.471356License: CC BY

Fault Tolerant Training for Optimal Interpolative NetsFinal published version, 1.62 MBLicense: CC BY

Cite this

@article{3982c5d64350439caa2815a76226bd9d,

title = "Fault Tolerant Training for Optimal Interpolative Nets",

abstract = " The optimal interpolative (OI) classification network is extended to include fault tolerance and make the network more robust to the loss of a neuron. The OI net has the characteristic that the training data are fit with no more neurons than necessary. Fault tolerance further reduces the number of neurons generated during the learning procedure while maintaining the generalization capabilities of the network. The learning algorithm for the fault-tolerant OI net is presented in a recursive formal, allowing for relatively short training times. A simulated fault-tolerant OI net is tested on a navigation satellite selection problem",

keywords = "Classification network, Fault-tolerant training, Generalization, Learning procedure, Navigation satellite selection, Neural nets, Optimal interpolative nets",

author = "Simon, \{Daniel J.\} and Hossny El-Sherief",

note = "D. Simon and H. El-Sherief. (1995). Fault Tolerant Training for Optimal Interpolative Nets, IEEE Transactions on Neural Networks, 6(6), 1531-1535, doi: 10.1109/72.471356.",

year = "1995",

month = nov,

day = "1",

doi = "10.1109/72.471356",

language = "American English",

volume = "6",

journal = "IEEE Transactions on Neural Networks",

}

TY - JOUR

T1 - Fault Tolerant Training for Optimal Interpolative Nets

AU - Simon, Daniel J.

AU - El-Sherief, Hossny

N1 - D. Simon and H. El-Sherief. (1995). Fault Tolerant Training for Optimal Interpolative Nets, IEEE Transactions on Neural Networks, 6(6), 1531-1535, doi: 10.1109/72.471356.

PY - 1995/11/1

Y1 - 1995/11/1

N2 - The optimal interpolative (OI) classification network is extended to include fault tolerance and make the network more robust to the loss of a neuron. The OI net has the characteristic that the training data are fit with no more neurons than necessary. Fault tolerance further reduces the number of neurons generated during the learning procedure while maintaining the generalization capabilities of the network. The learning algorithm for the fault-tolerant OI net is presented in a recursive formal, allowing for relatively short training times. A simulated fault-tolerant OI net is tested on a navigation satellite selection problem

AB - The optimal interpolative (OI) classification network is extended to include fault tolerance and make the network more robust to the loss of a neuron. The OI net has the characteristic that the training data are fit with no more neurons than necessary. Fault tolerance further reduces the number of neurons generated during the learning procedure while maintaining the generalization capabilities of the network. The learning algorithm for the fault-tolerant OI net is presented in a recursive formal, allowing for relatively short training times. A simulated fault-tolerant OI net is tested on a navigation satellite selection problem

KW - Classification network

KW - Fault-tolerant training

KW - Generalization

KW - Learning procedure

KW - Navigation satellite selection

KW - Neural nets

KW - Optimal interpolative nets

UR - https://engagedscholarship.csuohio.edu/enece_facpub/153

U2 - 10.1109/72.471356

DO - 10.1109/72.471356

M3 - Article

VL - 6

JO - IEEE Transactions on Neural Networks

JF - IEEE Transactions on Neural Networks

ER -

Fault Tolerant Training for Optimal Interpolative Nets

Abstract

Keywords

Disciplines

Access to Document

Other files and links

Cite this