Emerging memory technologies for neuromorphic computing

Chul Heung Kim; Suhwan Lim; Sung Yun Woo; Won Mook Kang; Young Tak Seo; Sung Tae Lee; Soochang Lee; Dongseok Kwon; Seongbin Oh; Yoohyun Noh; Hyeongsu Kim; Jangsaeng Kim; Jong Ho Bae; Jong Ho Lee

doi:10.1088/1361-6528/aae975

Emerging memory technologies for neuromorphic computing

Chul Heung Kim, Suhwan Lim, Sung Yun Woo, Won Mook Kang, Young Tak Seo, Sung Tae Lee, Soochang Lee, Dongseok Kwon, Seongbin Oh, Yoohyun Noh, Hyeongsu Kim, Jangsaeng Kim, Jong Ho Bae, Jong Ho Lee

School of Electronics Engineering

Seoul National University

Research output: Contribution to journal › Review article › peer-review

66 Scopus citations

Abstract

In this paper, we reviewed the recent trends on neuromorphic computing using emerging memory technologies. Two representative learning algorithms used to implement a hardware-based neural network are described as a bio-inspired learning algorithm and software-based learning algorithm, in particular back-propagation. The requirements of the synaptic device to apply each algorithm were analyzed. Then, we reviewed the research trends of synaptic devices to implement an artificial neural network.

Original language	English
Article number	032001
Journal	Nanotechnology
Volume	30
Issue number	3
DOIs	https://doi.org/10.1088/1361-6528/aae975
State	Published - 18 Jan 2019

Keywords

back-propagation (BP)
emerging memory
neuromorphic computing
spike-rate-dependent plasticity (SRDP)
spike-timing-dependent plasticity (STDP)
synaptic device

Access to Document

10.1088/1361-6528/aae975

Cite this

@article{f3505d1008ed4ff28aa0fc6ec4470ecd,

title = "Emerging memory technologies for neuromorphic computing",

abstract = "In this paper, we reviewed the recent trends on neuromorphic computing using emerging memory technologies. Two representative learning algorithms used to implement a hardware-based neural network are described as a bio-inspired learning algorithm and software-based learning algorithm, in particular back-propagation. The requirements of the synaptic device to apply each algorithm were analyzed. Then, we reviewed the research trends of synaptic devices to implement an artificial neural network.",

keywords = "back-propagation (BP), emerging memory, neuromorphic computing, spike-rate-dependent plasticity (SRDP), spike-timing-dependent plasticity (STDP), synaptic device",

author = "Kim, {Chul Heung} and Suhwan Lim and Woo, {Sung Yun} and Kang, {Won Mook} and Seo, {Young Tak} and Lee, {Sung Tae} and Soochang Lee and Dongseok Kwon and Seongbin Oh and Yoohyun Noh and Hyeongsu Kim and Jangsaeng Kim and Bae, {Jong Ho} and Lee, {Jong Ho}",

note = "Publisher Copyright: {\textcopyright} 2018 IOP Publishing Ltd.",

year = "2019",

month = jan,

day = "18",

doi = "10.1088/1361-6528/aae975",

language = "English",

volume = "30",

journal = "Nanotechnology",

issn = "0957-4484",

publisher = "IOP Publishing Ltd.",

number = "3",

}

TY - JOUR

T1 - Emerging memory technologies for neuromorphic computing

AU - Kim, Chul Heung

AU - Lim, Suhwan

AU - Woo, Sung Yun

AU - Kang, Won Mook

AU - Seo, Young Tak

AU - Lee, Sung Tae

AU - Lee, Soochang

AU - Kwon, Dongseok

AU - Oh, Seongbin

AU - Noh, Yoohyun

AU - Kim, Hyeongsu

AU - Kim, Jangsaeng

AU - Bae, Jong Ho

AU - Lee, Jong Ho

PY - 2019/1/18

Y1 - 2019/1/18

N2 - In this paper, we reviewed the recent trends on neuromorphic computing using emerging memory technologies. Two representative learning algorithms used to implement a hardware-based neural network are described as a bio-inspired learning algorithm and software-based learning algorithm, in particular back-propagation. The requirements of the synaptic device to apply each algorithm were analyzed. Then, we reviewed the research trends of synaptic devices to implement an artificial neural network.

AB - In this paper, we reviewed the recent trends on neuromorphic computing using emerging memory technologies. Two representative learning algorithms used to implement a hardware-based neural network are described as a bio-inspired learning algorithm and software-based learning algorithm, in particular back-propagation. The requirements of the synaptic device to apply each algorithm were analyzed. Then, we reviewed the research trends of synaptic devices to implement an artificial neural network.

KW - back-propagation (BP)

KW - emerging memory

KW - neuromorphic computing

KW - spike-rate-dependent plasticity (SRDP)

KW - spike-timing-dependent plasticity (STDP)

KW - synaptic device

UR - http://www.scopus.com/inward/record.url?scp=85056696666&partnerID=8YFLogxK

U2 - 10.1088/1361-6528/aae975

DO - 10.1088/1361-6528/aae975

M3 - Review article

C2 - 30422812

AN - SCOPUS:85056696666

SN - 0957-4484

VL - 30

JO - Nanotechnology

JF - Nanotechnology

IS - 3

M1 - 032001

ER -

Emerging memory technologies for neuromorphic computing

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this