Spiking neural networks based on two-dimensional materials

Juan B. Roldan; David Maldonadoep; Cristina Aguilera-Pedregosa; Enrique Moreno; Fernando Aguirre; Rocio Romero-Zaliz; Angel M. Garcia-Vico; Yaqing Shen; Mario Lanza

doi:10.1038/s41699-022-00341-5

Spiking neural networks based on two-dimensional materials

Juan B. Roldan, David Maldonadoep, Cristina Aguilera-Pedregosa, Enrique Moreno, Fernando Aguirre, Rocio Romero-Zaliz, Angel M. Garcia-Vico, Yaqing Shen, Mario Lanza

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

26 Scopus citations

Abstract

The development of artificial neural networks using memristors is gaining a lot of interest among technological companies because it can reduce the computing time and energy consumption. There is still no memristor, made of any material, capable to provide the ideal figures-of-merit required for the implementation of artificial neural networks, meaning that more research is required. Here we present the use of multilayer hexagonal boron nitride based memristors to implement spiking neural networks for image classification. Our study indicates that the recognition accuracy of the network is high, and that can be resilient to device variability if the number of neurons employed is large enough. There are very few studies that present the use of a two-dimensional material for the implementation of synapses of different features; in our case, in addition to a study of the synaptic characteristics of our memristive devices, we deal with complete spiking neural network training and inference processes.

Original language	English (US)
Journal	npj 2D Materials and Applications
Volume	6
Issue number	1
DOIs	https://doi.org/10.1038/s41699-022-00341-5
State	Published - Sep 9 2022

Bibliographical note

KAUST Repository Item: Exported on 2022-09-19
Acknowledgements: This work has been supported by the Ministry of Science and Technology of China (2018YFE0100800), the National Natural Science Foundation of China (grant no. 61874075), the Collaborative Innovation Centre of Suzhou Nano Science & Technology, the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the 111 Project from the State Administration of Foreign Experts Affairs of China. We also acknowledge the Consejería de Conocimiento, Investigación y Universidad, Junta de Andalucía (Spain) and European Regional Development Fund (ERDF) under projects A-TIC-117-UGR18, B-TIC-624-UGR20 and IE2017-5414, as well as the Spanish Ministry of Science, Innovation and Universities and ERDF fund under project RTI2018-098983-B-I00. M.L. acknowledges generous support from the King Abdullah University of Science and Technology.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1038/s41699-022-00341-5

https://repository.kaust.edu.sa/bitstream/10754/681554/1/s41699-022-00341-5.pdf

Cite this

@article{e9671fc890a34194b428e97e5bbdb88c,

title = "Spiking neural networks based on two-dimensional materials",

abstract = "The development of artificial neural networks using memristors is gaining a lot of interest among technological companies because it can reduce the computing time and energy consumption. There is still no memristor, made of any material, capable to provide the ideal figures-of-merit required for the implementation of artificial neural networks, meaning that more research is required. Here we present the use of multilayer hexagonal boron nitride based memristors to implement spiking neural networks for image classification. Our study indicates that the recognition accuracy of the network is high, and that can be resilient to device variability if the number of neurons employed is large enough. There are very few studies that present the use of a two-dimensional material for the implementation of synapses of different features; in our case, in addition to a study of the synaptic characteristics of our memristive devices, we deal with complete spiking neural network training and inference processes.",

author = "Roldan, {Juan B.} and David Maldonadoep and Cristina Aguilera-Pedregosa and Enrique Moreno and Fernando Aguirre and Rocio Romero-Zaliz and Garcia-Vico, {Angel M.} and Yaqing Shen and Mario Lanza",

note = "KAUST Repository Item: Exported on 2022-09-19 Acknowledgements: This work has been supported by the Ministry of Science and Technology of China (2018YFE0100800), the National Natural Science Foundation of China (grant no. 61874075), the Collaborative Innovation Centre of Suzhou Nano Science & Technology, the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the 111 Project from the State Administration of Foreign Experts Affairs of China. We also acknowledge the Consejer{\'i}a de Conocimiento, Investigaci{\'o}n y Universidad, Junta de Andaluc{\'i}a (Spain) and European Regional Development Fund (ERDF) under projects A-TIC-117-UGR18, B-TIC-624-UGR20 and IE2017-5414, as well as the Spanish Ministry of Science, Innovation and Universities and ERDF fund under project RTI2018-098983-B-I00. M.L. acknowledges generous support from the King Abdullah University of Science and Technology.",

year = "2022",

month = sep,

day = "9",

doi = "10.1038/s41699-022-00341-5",

language = "English (US)",

volume = "6",

journal = "npj 2D Materials and Applications",

issn = "2397-7132",

publisher = "Nature Research",

number = "1",

}

TY - JOUR

T1 - Spiking neural networks based on two-dimensional materials

AU - Roldan, Juan B.

AU - Maldonadoep, David

AU - Aguilera-Pedregosa, Cristina

AU - Moreno, Enrique

AU - Aguirre, Fernando

AU - Romero-Zaliz, Rocio

AU - Garcia-Vico, Angel M.

AU - Shen, Yaqing

AU - Lanza, Mario

N1 - KAUST Repository Item: Exported on 2022-09-19 Acknowledgements: This work has been supported by the Ministry of Science and Technology of China (2018YFE0100800), the National Natural Science Foundation of China (grant no. 61874075), the Collaborative Innovation Centre of Suzhou Nano Science & Technology, the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the 111 Project from the State Administration of Foreign Experts Affairs of China. We also acknowledge the Consejería de Conocimiento, Investigación y Universidad, Junta de Andalucía (Spain) and European Regional Development Fund (ERDF) under projects A-TIC-117-UGR18, B-TIC-624-UGR20 and IE2017-5414, as well as the Spanish Ministry of Science, Innovation and Universities and ERDF fund under project RTI2018-098983-B-I00. M.L. acknowledges generous support from the King Abdullah University of Science and Technology.

PY - 2022/9/9

Y1 - 2022/9/9

N2 - The development of artificial neural networks using memristors is gaining a lot of interest among technological companies because it can reduce the computing time and energy consumption. There is still no memristor, made of any material, capable to provide the ideal figures-of-merit required for the implementation of artificial neural networks, meaning that more research is required. Here we present the use of multilayer hexagonal boron nitride based memristors to implement spiking neural networks for image classification. Our study indicates that the recognition accuracy of the network is high, and that can be resilient to device variability if the number of neurons employed is large enough. There are very few studies that present the use of a two-dimensional material for the implementation of synapses of different features; in our case, in addition to a study of the synaptic characteristics of our memristive devices, we deal with complete spiking neural network training and inference processes.

AB - The development of artificial neural networks using memristors is gaining a lot of interest among technological companies because it can reduce the computing time and energy consumption. There is still no memristor, made of any material, capable to provide the ideal figures-of-merit required for the implementation of artificial neural networks, meaning that more research is required. Here we present the use of multilayer hexagonal boron nitride based memristors to implement spiking neural networks for image classification. Our study indicates that the recognition accuracy of the network is high, and that can be resilient to device variability if the number of neurons employed is large enough. There are very few studies that present the use of a two-dimensional material for the implementation of synapses of different features; in our case, in addition to a study of the synaptic characteristics of our memristive devices, we deal with complete spiking neural network training and inference processes.

UR - http://hdl.handle.net/10754/681554

UR - https://www.nature.com/articles/s41699-022-00341-5

U2 - 10.1038/s41699-022-00341-5

DO - 10.1038/s41699-022-00341-5

M3 - Article

SN - 2397-7132

VL - 6

JO - npj 2D Materials and Applications

JF - npj 2D Materials and Applications

IS - 1

ER -

Spiking neural networks based on two-dimensional materials

Abstract

Bibliographical note

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this