Neural networks update

Abstract

Summary form only given, as follows. Neural networks have been intensively studied as a discipline in their own right in the last five years (late 1980s, early 1990s). Initial claims were extremely ambitious; by using the brain's computing principles, networks would eliminate programming, revolutionize computer architecture and sensor interfacing, make analog VLSI a reality, and give guidance to a new understanding of human cognition. Work in two areas is described: statistical methods to deal with classification, prediction, and control in data-rich, intuition-poor problems; and VLSI solutions, both in digital and analog styles, to accommodate these architectures.<>