Memristors with Initial Low‐Resistive State for Efficient Neuromorphic Systems

Kaichen Zhu; Mohammad Reza Mahmoodi; Zahra Fahimi; Yiping Xiao; Tao Wang; Kristýna Bukvišová; Miroslav Kolíbal; Juan B. Roldan; David Perez; Fernando Aguirre; Mario Lanza

doi:10.1002/aisy.202200001

Memristors with Initial Low‐Resistive State for Efficient Neuromorphic Systems

Kaichen Zhu, Mohammad Reza Mahmoodi, Zahra Fahimi, Yiping Xiao, Tao Wang, Kristýna Bukvišová, Miroslav Kolíbal, Juan B. Roldan, David Perez, Fernando Aguirre, Mario Lanza

Materials Science and Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Memristive electronic synapses are attractive to construct artificial neural networks (ANNs) for neuromorphic computing systems, owing to their excellent electronic performance, high integration density, and low cost. However, the necessity of initializing their conductance through a forming process requires additional peripheral hardware and complex programming algorithms. Herein, the first fabrication of memristors that are initially in low-resistive state (LRS) is reported, which exhibit homogenous initial resistance and switching voltages. When used as electronic synapses in a neuromorphic system to classify images from the CIFAR-10 dataset (Canadian Institute For Advanced Research), the memristors offer ×1.83 better throughput per area and consume ×0.85 less energy than standard memristors (i.e., with the necessity of forming), which stems from ≈63% better density and ≈17% faster operation. It is demonstrated in the results that tuning the local properties of materials embedded in memristive electronic synapses is an attractive strategy that can lead to an improved neuromorphic performance at the system level.

Original language	English
Journal	Advanced Intelligent Systems
DOIs	https://doi.org/10.1002/aisy.202200001
State	Published - 21 Mar 2022

Access to Document

10.1002/aisy.202200001

Cite this

@article{9e3dc1ecb8c64755b20fd37bc618cc38,

title = "Memristors with Initial Low‐Resistive State for Efficient Neuromorphic Systems",

abstract = "Memristive electronic synapses are attractive to construct artificial neural networks (ANNs) for neuromorphic computing systems, owing to their excellent electronic performance, high integration density, and low cost. However, the necessity of initializing their conductance through a forming process requires additional peripheral hardware and complex programming algorithms. Herein, the first fabrication of memristors that are initially in low-resistive state (LRS) is reported, which exhibit homogenous initial resistance and switching voltages. When used as electronic synapses in a neuromorphic system to classify images from the CIFAR-10 dataset (Canadian Institute For Advanced Research), the memristors offer ×1.83 better throughput per area and consume ×0.85 less energy than standard memristors (i.e., with the necessity of forming), which stems from ≈63% better density and ≈17% faster operation. It is demonstrated in the results that tuning the local properties of materials embedded in memristive electronic synapses is an attractive strategy that can lead to an improved neuromorphic performance at the system level.",

author = "Kaichen Zhu and Mahmoodi, {Mohammad Reza} and Zahra Fahimi and Yiping Xiao and Tao Wang and Krist{\'y}na Bukvi{\v s}ov{\'a} and Miroslav Kol{\'i}bal and Roldan, {Juan B.} and David Perez and Fernando Aguirre and Mario Lanza",

year = "2022",

month = mar,

day = "21",

doi = "10.1002/aisy.202200001",

language = "英语",

journal = "Advanced Intelligent Systems",

issn = "2640-4567",

}

Memristors with Initial Low‐Resistive State for Efficient Neuromorphic Systems. / Zhu, Kaichen; Mahmoodi, Mohammad Reza; Fahimi, Zahra; Xiao, Yiping; Wang, Tao; Bukvišová, Kristýna; Kolíbal, Miroslav; Roldan, Juan B.; Perez, David; Aguirre, Fernando; Lanza, Mario.

In: Advanced Intelligent Systems, 21.03.2022.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - Memristors with Initial Low‐Resistive State for Efficient Neuromorphic Systems

AU - Zhu, Kaichen

AU - Mahmoodi, Mohammad Reza

AU - Fahimi, Zahra

AU - Xiao, Yiping

AU - Wang, Tao

AU - Bukvišová, Kristýna

AU - Kolíbal, Miroslav

AU - Roldan, Juan B.

AU - Perez, David

AU - Aguirre, Fernando

AU - Lanza, Mario

PY - 2022/3/21

Y1 - 2022/3/21

N2 - Memristive electronic synapses are attractive to construct artificial neural networks (ANNs) for neuromorphic computing systems, owing to their excellent electronic performance, high integration density, and low cost. However, the necessity of initializing their conductance through a forming process requires additional peripheral hardware and complex programming algorithms. Herein, the first fabrication of memristors that are initially in low-resistive state (LRS) is reported, which exhibit homogenous initial resistance and switching voltages. When used as electronic synapses in a neuromorphic system to classify images from the CIFAR-10 dataset (Canadian Institute For Advanced Research), the memristors offer ×1.83 better throughput per area and consume ×0.85 less energy than standard memristors (i.e., with the necessity of forming), which stems from ≈63% better density and ≈17% faster operation. It is demonstrated in the results that tuning the local properties of materials embedded in memristive electronic synapses is an attractive strategy that can lead to an improved neuromorphic performance at the system level.

AB - Memristive electronic synapses are attractive to construct artificial neural networks (ANNs) for neuromorphic computing systems, owing to their excellent electronic performance, high integration density, and low cost. However, the necessity of initializing their conductance through a forming process requires additional peripheral hardware and complex programming algorithms. Herein, the first fabrication of memristors that are initially in low-resistive state (LRS) is reported, which exhibit homogenous initial resistance and switching voltages. When used as electronic synapses in a neuromorphic system to classify images from the CIFAR-10 dataset (Canadian Institute For Advanced Research), the memristors offer ×1.83 better throughput per area and consume ×0.85 less energy than standard memristors (i.e., with the necessity of forming), which stems from ≈63% better density and ≈17% faster operation. It is demonstrated in the results that tuning the local properties of materials embedded in memristive electronic synapses is an attractive strategy that can lead to an improved neuromorphic performance at the system level.

U2 - 10.1002/aisy.202200001

DO - 10.1002/aisy.202200001

M3 - 文章

JO - Advanced Intelligent Systems

JF - Advanced Intelligent Systems

SN - 2640-4567

ER -

Memristors with Initial Low‐Resistive State for Efficient Neuromorphic Systems

Abstract

Access to Document

Fingerprint

Cite this