Date of Award
Spring 1997
Document Type
Thesis - Restricted
Degree Name
Master of Science (MS)
Department
Computer Science
First Advisor
Belfore, Lee A.
Second Advisor
Heinen, James
Third Advisor
Feng, Xing
Abstract
This work was done in an attempt to fill a need in the manufacture of WSI and VLSI Artificial Neural Networks (ANN). That need is the development of efficient test procedures for testing WSI and VLSI ANN implementations. As will be shown in this work, the current test procedures are limited in their coverage of the ANN components or are limited to specific types of ANN architectures. The Low Activation Gain Fault Detection (LAGFD) test algorithm presented in this work assumes no particular network architecture. The only requirements of the algorithm is that the ANN interconnection matrix be programmable and fully connectable, and that the activation function of each neuron be programmable to a low value, i.e. G * u < l. The LAGFD algorithm is shown to be implementable as a Built-in Self Test (BIST) to be included directly on the silicon ANN implementation
Recommended Citation
Fleischer, Curtis A., "A Neural Network Test Methodology for Networks with Programmable Low-Gain Activation Functions" (1997). Master's Theses (1922-2009) Access restricted to Marquette Campus. 4582.
https://epublications.marquette.edu/theses/4582