NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks

Abstract Artificial Intelligence (AI) at the edge has become a hot subject of the recent technology-minded publications. The challenges related to IoT nodes gave rise to research on efficient hardware-based accelerators. In this context, analog memristor devices are crucial elements to efficiently p...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Heba Abunahla, Yasmin Halawani, Anas Alazzam, Baker Mohammad
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2020
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/9f317778f3d04fb5badbe463b2cb6632
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:9f317778f3d04fb5badbe463b2cb6632
record_format	dspace
spelling	oai:doaj.org-article:9f317778f3d04fb5badbe463b2cb66322021-12-02T17:52:33ZNeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks10.1038/s41598-020-66413-y2045-2322https://doaj.org/article/9f317778f3d04fb5badbe463b2cb66322020-06-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-66413-yhttps://doaj.org/toc/2045-2322Abstract Artificial Intelligence (AI) at the edge has become a hot subject of the recent technology-minded publications. The challenges related to IoT nodes gave rise to research on efficient hardware-based accelerators. In this context, analog memristor devices are crucial elements to efficiently perform the multiply-and-add (MAD) operations found in many AI algorithms. This is due to the ability of memristor devices to perform in-memory-computing (IMC) in a way that mimics the synapses in human brain. Here, we present a novel planar analog memristor, namely NeuroMem, that includes a partially reduced Graphene Oxide (prGO) thin film. The analog and non-volatile resistance switching of NeuroMem enable tuning it to any value within the RON and ROFF range. These two features make NeuroMem a potential candidate for emerging IMC applications such as inference engine for AI systems. Moreover, the prGO thin film of the memristor is patterned on a flexible substrate of Cyclic Olefin Copolymer (COC) using standard microfabrication techniques. This provides new opportunities for simple, flexible, and cost-effective fabrication of solution-based Graphene-based memristors. In addition to providing detailed electrical characterization of the device, a crossbar of the technology has been fabricated to demonstrate its ability to implement IMC for MAD operations targeting fully connected layer of Artificial Neural Network. This work is the first to report on the great potential of this technology for AI inference application especially for edge devices.Heba AbunahlaYasmin HalawaniAnas AlazzamBaker MohammadNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 10, Iss 1, Pp 1-11 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Heba Abunahla Yasmin Halawani Anas Alazzam Baker Mohammad NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks
description	Abstract Artificial Intelligence (AI) at the edge has become a hot subject of the recent technology-minded publications. The challenges related to IoT nodes gave rise to research on efficient hardware-based accelerators. In this context, analog memristor devices are crucial elements to efficiently perform the multiply-and-add (MAD) operations found in many AI algorithms. This is due to the ability of memristor devices to perform in-memory-computing (IMC) in a way that mimics the synapses in human brain. Here, we present a novel planar analog memristor, namely NeuroMem, that includes a partially reduced Graphene Oxide (prGO) thin film. The analog and non-volatile resistance switching of NeuroMem enable tuning it to any value within the RON and ROFF range. These two features make NeuroMem a potential candidate for emerging IMC applications such as inference engine for AI systems. Moreover, the prGO thin film of the memristor is patterned on a flexible substrate of Cyclic Olefin Copolymer (COC) using standard microfabrication techniques. This provides new opportunities for simple, flexible, and cost-effective fabrication of solution-based Graphene-based memristors. In addition to providing detailed electrical characterization of the device, a crossbar of the technology has been fabricated to demonstrate its ability to implement IMC for MAD operations targeting fully connected layer of Artificial Neural Network. This work is the first to report on the great potential of this technology for AI inference application especially for edge devices.
format	article
author	Heba Abunahla Yasmin Halawani Anas Alazzam Baker Mohammad
author_facet	Heba Abunahla Yasmin Halawani Anas Alazzam Baker Mohammad
author_sort	Heba Abunahla
title	NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks
title_short	NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks
title_full	NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks
title_fullStr	NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks
title_full_unstemmed	NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks
title_sort	neuromem: analog graphene-based resistive memory for artificial neural networks
publisher	Nature Portfolio
publishDate	2020
url	https://doaj.org/article/9f317778f3d04fb5badbe463b2cb6632
work_keys_str_mv	AT hebaabunahla neuromemanaloggraphenebasedresistivememoryforartificialneuralnetworks AT yasminhalawani neuromemanaloggraphenebasedresistivememoryforartificialneuralnetworks AT anasalazzam neuromemanaloggraphenebasedresistivememoryforartificialneuralnetworks AT bakermohammad neuromemanaloggraphenebasedresistivememoryforartificialneuralnetworks
_version_	1718379183148630016

NeuroMem: Analog Graphene-Based Resistive Memory for Artificial Neural Networks

Ejemplares similares