Publication Details
Author: R. D. Clay and C. H. Sequin
Group: ICSI Technical Reports
Date: April 1991
PDF: http://www.icsi.berkeley.edu/pubs/techreports/tr-91-025.pdf
Overview:
The worst case output errors produced by the failure of a hidden neuron in layered feed-forward ANNs are investigated. These errors can be much worse than simply the loss of the contribution of a neuron whose output goes to zero. A much larger erroneous signal can be produced when the failure sets the value of the hidden neuron to one of the power supply voltages.
A new method is investigated that limits the fractional error in the output signal of a feed-forward net due to such saturated hidden unit faults in analog function approximation tasks. The number of hidden units is significantly increased, and the maximal contribution of each unit is limited to a small fraction of the net output signal. To achieve a large localized output signal, several Gaussian hidden units are moved into the same location in the input domain and the gain of the linear summing output unit is suitably adjusted. Since the contribution of each unit is equal in magnitude, there is only a modest error under any possible failure mode.
Bibliographic Information:
ICSI Technical Report TR-91-025
Bibliographic Reference:
R. D. Clay and C. H. Sequin. Limiting Fault-Induced Output Errors In ANNs. ICSI Technical Report TR-91-025, April 1991