Abstract | ||
---|---|---|
We develop a neural network that learns to separate the nominal from the faulty instances of a circuit in a measurement space. We demonstrate that the required separation boundaries are, in general, non-linear. Unlike previous solutions which draw hyperplanes, our network is capable of drawing the necessary non-linear hypersurfaces. The hypersurfaces translate to test criteria that are strongly correlated to functional tests. A feature selection algorithm interacts with the network to identify a discriminative low-dimensional measurement space. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1145/1057661.1057684 | ACM Great Lakes Symposium on VLSI |
Keywords | Field | DocType |
analog measurement space,previous solution,required separation boundary,generating decision region,measurement space,neural network,faulty instance,necessary non-linear hypersurfaces,discriminative low-dimensional measurement space,functional test,feature selection algorithm interacts,feature selection,functional testing,neural networks,analog circuits | Analogue electronics,Feature selection,Computer science,Algorithm,Electronic engineering,Theoretical computer science,Hyperplane,Artificial neural network,Discriminative model | Conference |
ISBN | Citations | PageRank |
1-59593-057-4 | 1 | 0.36 |
References | Authors | |
5 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Haralampos-G. D. Stratigopoulos | 1 | 252 | 28.06 |
Yiorgos Makris | 2 | 1365 | 107.21 |