A low energy oxide-based electronic synaptic device for neuromorphic visual systems with tolerance to device variation

Shimeng Yu, Bin Gao, Zheng Fang, Hongyu Yu, Jinfeng Kang, H. S Philip Wong

Research output: Contribution to journalArticle

240 Citations (Scopus)

Abstract

Neuromorphic computing is an emerging computing paradigm beyond the conventional digital von Neumann computation. An oxide-based resistive switching memory is engineered to emulate synaptic devices. At the device level, the gradual resistance modulation is characterized by hundreds of identical pulses, achieving a low energy consumption of less than 1 pJ per spike. Furthermore, a stochastic compact model is developed to quantify the device switching dynamics and variation. At system level, the performance of an artificial visual system on the image orientation or edge detection with 16 348 oxide-based synaptic devices is simulated, successfully demonstrating a key feature of neuromorphic computing: tolerance to device variation.

Original languageEnglish (US)
Pages (from-to)1774-1779
Number of pages6
JournalAdvanced Materials
Volume25
Issue number12
DOIs
StatePublished - Mar 25 2013
Externally publishedYes

Fingerprint

Oxides
Edge detection
Energy utilization
Modulation
Data storage equipment

Keywords

  • artificial visual systems
  • neuromorphic computing
  • oxide RRAM
  • resistive switching
  • synaptic devices

ASJC Scopus subject areas

  • Materials Science(all)
  • Mechanics of Materials
  • Mechanical Engineering

Cite this

A low energy oxide-based electronic synaptic device for neuromorphic visual systems with tolerance to device variation. / Yu, Shimeng; Gao, Bin; Fang, Zheng; Yu, Hongyu; Kang, Jinfeng; Wong, H. S Philip.

In: Advanced Materials, Vol. 25, No. 12, 25.03.2013, p. 1774-1779.

Research output: Contribution to journalArticle

Yu, Shimeng ; Gao, Bin ; Fang, Zheng ; Yu, Hongyu ; Kang, Jinfeng ; Wong, H. S Philip. / A low energy oxide-based electronic synaptic device for neuromorphic visual systems with tolerance to device variation. In: Advanced Materials. 2013 ; Vol. 25, No. 12. pp. 1774-1779.
@article{a7a4492dc59a4bb7afbdd78404eb1a29,
title = "A low energy oxide-based electronic synaptic device for neuromorphic visual systems with tolerance to device variation",
abstract = "Neuromorphic computing is an emerging computing paradigm beyond the conventional digital von Neumann computation. An oxide-based resistive switching memory is engineered to emulate synaptic devices. At the device level, the gradual resistance modulation is characterized by hundreds of identical pulses, achieving a low energy consumption of less than 1 pJ per spike. Furthermore, a stochastic compact model is developed to quantify the device switching dynamics and variation. At system level, the performance of an artificial visual system on the image orientation or edge detection with 16 348 oxide-based synaptic devices is simulated, successfully demonstrating a key feature of neuromorphic computing: tolerance to device variation.",
keywords = "artificial visual systems, neuromorphic computing, oxide RRAM, resistive switching, synaptic devices",
author = "Shimeng Yu and Bin Gao and Zheng Fang and Hongyu Yu and Jinfeng Kang and Wong, {H. S Philip}",
year = "2013",
month = "3",
day = "25",
doi = "10.1002/adma.201203680",
language = "English (US)",
volume = "25",
pages = "1774--1779",
journal = "Advanced Materials",
issn = "0935-9648",
publisher = "Wiley-VCH Verlag",
number = "12",

}

TY - JOUR

T1 - A low energy oxide-based electronic synaptic device for neuromorphic visual systems with tolerance to device variation

AU - Yu, Shimeng

AU - Gao, Bin

AU - Fang, Zheng

AU - Yu, Hongyu

AU - Kang, Jinfeng

AU - Wong, H. S Philip

PY - 2013/3/25

Y1 - 2013/3/25

N2 - Neuromorphic computing is an emerging computing paradigm beyond the conventional digital von Neumann computation. An oxide-based resistive switching memory is engineered to emulate synaptic devices. At the device level, the gradual resistance modulation is characterized by hundreds of identical pulses, achieving a low energy consumption of less than 1 pJ per spike. Furthermore, a stochastic compact model is developed to quantify the device switching dynamics and variation. At system level, the performance of an artificial visual system on the image orientation or edge detection with 16 348 oxide-based synaptic devices is simulated, successfully demonstrating a key feature of neuromorphic computing: tolerance to device variation.

AB - Neuromorphic computing is an emerging computing paradigm beyond the conventional digital von Neumann computation. An oxide-based resistive switching memory is engineered to emulate synaptic devices. At the device level, the gradual resistance modulation is characterized by hundreds of identical pulses, achieving a low energy consumption of less than 1 pJ per spike. Furthermore, a stochastic compact model is developed to quantify the device switching dynamics and variation. At system level, the performance of an artificial visual system on the image orientation or edge detection with 16 348 oxide-based synaptic devices is simulated, successfully demonstrating a key feature of neuromorphic computing: tolerance to device variation.

KW - artificial visual systems

KW - neuromorphic computing

KW - oxide RRAM

KW - resistive switching

KW - synaptic devices

UR - http://www.scopus.com/inward/record.url?scp=84875158827&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875158827&partnerID=8YFLogxK

U2 - 10.1002/adma.201203680

DO - 10.1002/adma.201203680

M3 - Article

C2 - 23355110

AN - SCOPUS:84875158827

VL - 25

SP - 1774

EP - 1779

JO - Advanced Materials

JF - Advanced Materials

SN - 0935-9648

IS - 12

ER -