A Refreshable Analog VLSI Neural Network Chip with 400 Neurons and 40K Synapses

Yutaka Arima, Mitsuhiro Murasaki, Tsuyoshi Yamada, Atsushi Maeda, Hirofumi Shinohara

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

This paper describes an on-chip learning neural network LSI circuit that can refresh the analog storage synaptic weights located on the chip. The chip integrates 400 neurons and 40 000 synapses with a 0.8-μm double-poly double-metal CMOS technology. This device stores learned information by repeating the refresh process at 200-ms intervals.

Original languageEnglish
Pages (from-to)1854-1861
Number of pages8
JournalIEEE Journal of Solid-State Circuits
Volume27
Issue number12
DOIs
Publication statusPublished - 1992 Dec
Externally publishedYes

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'A Refreshable Analog VLSI Neural Network Chip with 400 Neurons and 40K Synapses'. Together they form a unique fingerprint.

  • Cite this