Abstract
The burgeoning interest in ‘neural networks’ (or ‘paralleldistributed processing’ or ‘connectionism’) is now an established fact. Whether this will grow into a new engineering discipline or not is still an open question. This article discusses the beginnings of a unified approach aimed at enabling engineers to design andbuild significant neural systems in an effort to improve the competence of computing systems in general.

This publication has 2 references indexed in Scilit: