Indium-Gallium-Zinc Oxide-Based Synaptic Charge Trap Flash for Spiking Neural Network-Restricted Boltzmann Machine

Eunpyo Park; Suyeon Jang; Gichang Noh; Yooyeon Jo; Dae Kyu Lee; In Soo Kim; Hyun Cheol Song; Sangbum Kim; Joon Young Kwak

doi:10.1021/acs.nanolett.3c03510

Indium-Gallium-Zinc Oxide-Based Synaptic Charge Trap Flash for Spiking Neural Network-Restricted Boltzmann Machine

Eunpyo Park, Suyeon Jang, Gichang Noh, Yooyeon Jo, Dae Kyu Lee, In Soo Kim, Hyun Cheol Song, Sangbum Kim, Joon Young Kwak

Semiconductor Engineering

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

5 Scopus citations

Abstract

Recently, neuromorphic computing has been proposed to overcome the drawbacks of the current von Neumann computing architecture. Especially, spiking neural network (SNN) has received significant attention due to its ability to mimic the spike-driven behavior of biological neurons and synapses, potentially leading to low-power consumption and other advantages. In this work, we designed the indium-gallium-zinc oxide (IGZO) channel charge-trap flash (CTF) synaptic device based on a HfO₂/Al₂O₃/Si₃N₄/Al₂O₃ layer. Our IGZO-based CTF device exhibits synaptic functions with 128 levels of synaptic weight states and spike-timing-dependent plasticity. The SNN-restricted Boltzmann machine was used to simulate the fabricated CTF device to evaluate the efficiency for the SNN system, achieving the high pattern-recognition accuracy of 83.9%. We believe that our results show the suitability of the fabricated IGZO CTF device as a synaptic device for neuromorphic computing.

Original language	English
Pages (from-to)	9626-9633
Number of pages	8
Journal	Nano Letters
Volume	23
Issue number	20
DOIs	https://doi.org/10.1021/acs.nanolett.3c03510
State	Published - 25 Oct 2023

Bibliographical note

Publisher Copyright:
© 2023 American Chemical Society.

Keywords

charge trap flash
neuromorphic computing
nonvolatile memory
oxide semiconductor
spiking neural network

Access to Document

10.1021/acs.nanolett.3c03510

Cite this

@article{42a71df4ec794fe3a1ec91b3b05fb56f,

title = "Indium-Gallium-Zinc Oxide-Based Synaptic Charge Trap Flash for Spiking Neural Network-Restricted Boltzmann Machine",

abstract = "Recently, neuromorphic computing has been proposed to overcome the drawbacks of the current von Neumann computing architecture. Especially, spiking neural network (SNN) has received significant attention due to its ability to mimic the spike-driven behavior of biological neurons and synapses, potentially leading to low-power consumption and other advantages. In this work, we designed the indium-gallium-zinc oxide (IGZO) channel charge-trap flash (CTF) synaptic device based on a HfO2/Al2O3/Si3N4/Al2O3 layer. Our IGZO-based CTF device exhibits synaptic functions with 128 levels of synaptic weight states and spike-timing-dependent plasticity. The SNN-restricted Boltzmann machine was used to simulate the fabricated CTF device to evaluate the efficiency for the SNN system, achieving the high pattern-recognition accuracy of 83.9%. We believe that our results show the suitability of the fabricated IGZO CTF device as a synaptic device for neuromorphic computing.",

keywords = "charge trap flash, neuromorphic computing, nonvolatile memory, oxide semiconductor, spiking neural network",

author = "Eunpyo Park and Suyeon Jang and Gichang Noh and Yooyeon Jo and Lee, {Dae Kyu} and Kim, {In Soo} and Song, {Hyun Cheol} and Sangbum Kim and Kwak, {Joon Young}",

note = "Publisher Copyright: {\textcopyright} 2023 American Chemical Society.",

year = "2023",

month = oct,

day = "25",

doi = "10.1021/acs.nanolett.3c03510",

language = "English",

volume = "23",

pages = "9626--9633",

journal = "Nano Letters",

issn = "1530-6984",

publisher = "American Chemical Society",

number = "20",

}

TY - JOUR

T1 - Indium-Gallium-Zinc Oxide-Based Synaptic Charge Trap Flash for Spiking Neural Network-Restricted Boltzmann Machine

AU - Park, Eunpyo

AU - Jang, Suyeon

AU - Noh, Gichang

AU - Jo, Yooyeon

AU - Lee, Dae Kyu

AU - Kim, In Soo

AU - Song, Hyun Cheol

AU - Kim, Sangbum

AU - Kwak, Joon Young

PY - 2023/10/25

Y1 - 2023/10/25

N2 - Recently, neuromorphic computing has been proposed to overcome the drawbacks of the current von Neumann computing architecture. Especially, spiking neural network (SNN) has received significant attention due to its ability to mimic the spike-driven behavior of biological neurons and synapses, potentially leading to low-power consumption and other advantages. In this work, we designed the indium-gallium-zinc oxide (IGZO) channel charge-trap flash (CTF) synaptic device based on a HfO2/Al2O3/Si3N4/Al2O3 layer. Our IGZO-based CTF device exhibits synaptic functions with 128 levels of synaptic weight states and spike-timing-dependent plasticity. The SNN-restricted Boltzmann machine was used to simulate the fabricated CTF device to evaluate the efficiency for the SNN system, achieving the high pattern-recognition accuracy of 83.9%. We believe that our results show the suitability of the fabricated IGZO CTF device as a synaptic device for neuromorphic computing.

AB - Recently, neuromorphic computing has been proposed to overcome the drawbacks of the current von Neumann computing architecture. Especially, spiking neural network (SNN) has received significant attention due to its ability to mimic the spike-driven behavior of biological neurons and synapses, potentially leading to low-power consumption and other advantages. In this work, we designed the indium-gallium-zinc oxide (IGZO) channel charge-trap flash (CTF) synaptic device based on a HfO2/Al2O3/Si3N4/Al2O3 layer. Our IGZO-based CTF device exhibits synaptic functions with 128 levels of synaptic weight states and spike-timing-dependent plasticity. The SNN-restricted Boltzmann machine was used to simulate the fabricated CTF device to evaluate the efficiency for the SNN system, achieving the high pattern-recognition accuracy of 83.9%. We believe that our results show the suitability of the fabricated IGZO CTF device as a synaptic device for neuromorphic computing.

KW - charge trap flash

KW - neuromorphic computing

KW - nonvolatile memory

KW - oxide semiconductor

KW - spiking neural network

UR - http://www.scopus.com/inward/record.url?scp=85175271496&partnerID=8YFLogxK

U2 - 10.1021/acs.nanolett.3c03510

DO - 10.1021/acs.nanolett.3c03510

M3 - Article

C2 - 37819875

AN - SCOPUS:85175271496

SN - 1530-6984

VL - 23

SP - 9626

EP - 9633

JO - Nano Letters

JF - Nano Letters

IS - 20

ER -

Indium-Gallium-Zinc Oxide-Based Synaptic Charge Trap Flash for Spiking Neural Network-Restricted Boltzmann Machine

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this