Implementation of Million Connections Neural Hardware with URAN-I

Autor: Ki-Hwan Ahn, Il Song Han
Rok vydání: 1993
Předmět:
Zdroj: ICANN ’93 ISBN: 9783540198390
DOI: 10.1007/978-1-4471-2063-6_307
Popis: This paper describes the neural hardware build of the full custom neural VLSI. The newly developed URAN-I neuro-chips of 135,424 connections are used to construct with their speed of hundreds giga connections per second. The basic hardware module is suggested with 8 chips and digital computer interface to yield thousands giga connections per second in speed and more than one million connections in size.
Databáze: OpenAIRE
načítá se...