English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Energy-efficient memcapacitor devices for neuromorphic computing

Demasius, K.-U., Kirschen, A., & Parkin, S. (2021). Energy-efficient memcapacitor devices for neuromorphic computing. Nature Electronics, 4, 748-756. doi:10.1038/s41928-021-00649-y.

Item is

Files

show Files
hide Files
:
s41928-021-00649-y.pdf (Publisher version), 5MB
Name:
s41928-021-00649-y.pdf
Description:
-
OA-Status:
Hybrid
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
2021
Copyright Info:
The Author(s)

Locators

show
hide
Locator:
https://doi.org/10.1038/s41928-021-00649-y (Publisher version)
Description:
-
OA-Status:
Hybrid

Creators

show
hide
 Creators:
Demasius, Kai-Uwe1, 2, Author
Kirschen, Aron3, Author
Parkin, Stuart1, Author                 
Affiliations:
1Nano-Systems from Ions, Spins and Electrons, Max Planck Institute of Microstructure Physics, Max Planck Society, ou_3287476              
2International Max Planck Research School for Science and Technology of Nano-Systems, Max Planck Institute of Microstructure Physics, Max Planck Society, Weinberg 2, 06120 Halle (Saale), Germany, ou_3399928              
3External Organizations, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: Data-intensive computing operations, such as training neural networks, are essential for applications in artificial intelligence but are energy intensive. One solution is to develop specialized hardware onto which neural networks can be directly mapped, and arrays of memristive devices can, for example, be trained to enable parallel multiply–accumulate operations. Here we show that memcapacitive devices that exploit the principle of charge shielding can offer a highly energy-efficient approach for implementing parallel multiply–accumulate operations. We fabricate a crossbar array of 156 microscale memcapacitor devices and use it to train a neural network that could distinguish the letters ‘M’, ‘P’ and ‘I’. Modelling these arrays suggests that this approach could offer an energy efficiency of 29,600 tera-operations per second per watt, while ensuring high precision (6–8 bits). Simulations also show that the devices could potentially be scaled down to a lateral size of around 45 nm.

Details

show
hide
Language(s):
 Dates: 2021-10-112021-10
 Publication Status: Issued
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: DOI: 10.1038/s41928-021-00649-y
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Nature Electronics
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: London : Springer Nature
Pages: - Volume / Issue: 4 Sequence Number: - Start / End Page: 748 - 756 Identifier: ISSN: 2520-1131
CoNE: https://pure.mpg.de/cone/journals/resource/25201131