Obtaining high precision operation from nonideal neural networks

Journal Article

Several potential neural architectures for an A/D converter are examined, and the level of nonidealities that can be tolerated by the network components without inhibiting high-precision operation through training is discussed. Behavioral-level simulations on sample converter networks with modeled nonidealities revealed a strong interrelationship between the network architectures and their tolerance to nonidealities.

Duke Authors

Cited Authors

  • Sculley, TL; Brooke, MA

Published Date

  • December 1, 1990

Published In

Volume / Issue

  • 3 /

Start / End Page

  • 1847 - 1850

International Standard Serial Number (ISSN)

  • 0271-4310

Citation Source

  • Scopus