A ferroelectric field effect transistor based synaptic weight cell

Matthew Jerry, Sourav Dutta, Arman Kazemi, Kai Ni, Jianchi Zhang, Pai Yu Chen, Pankaj Sharma, Shimeng Yu, X. Sharon Hu, Michael Niemier, Suman Datta

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Dense analog synaptic crossbar arrays are a promising candidate for neuromorphic hardware accelerators due to the ability to mitigate data movement by performing in-situ vector-matrix products and weight updates within the storage array itself. However, many analog weight storage cells suffer from long latencies or low dynamic ranges, limiting the achievable performance. In this work, we demonstrate that the voltage-controlled partial polarization switching dynamics in ferroelectric-field-effect transistors (FeFET) can be harnessed to enable a 32 state non-volatile analog synaptic weight cell with large dynamic range (67×) and low latency weight updates (50 ns) for an amplitude modulated pulse scheme.

Original languageEnglish (US)
Article number434001
JournalJournal of Physics D: Applied Physics
Volume51
Issue number43
DOIs
StatePublished - Aug 23 2018
Externally publishedYes

Fingerprint

Field effect transistors
Ferroelectric materials
field effect transistors
analogs
dynamic range
cells
low weight
Particle accelerators
hardware
accelerators
Polarization
Hardware
Electric potential
electric potential
polarization
products
pulses

Keywords

  • analog synapse
  • feld-effect-transistor
  • ferroelectric
  • Neuromorphic computing

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Acoustics and Ultrasonics
  • Surfaces, Coatings and Films

Cite this

Jerry, M., Dutta, S., Kazemi, A., Ni, K., Zhang, J., Chen, P. Y., ... Datta, S. (2018). A ferroelectric field effect transistor based synaptic weight cell. Journal of Physics D: Applied Physics, 51(43), [434001]. https://doi.org/10.1088/1361-6463/aad6f8

A ferroelectric field effect transistor based synaptic weight cell. / Jerry, Matthew; Dutta, Sourav; Kazemi, Arman; Ni, Kai; Zhang, Jianchi; Chen, Pai Yu; Sharma, Pankaj; Yu, Shimeng; Hu, X. Sharon; Niemier, Michael; Datta, Suman.

In: Journal of Physics D: Applied Physics, Vol. 51, No. 43, 434001, 23.08.2018.

Research output: Contribution to journalArticle

Jerry, M, Dutta, S, Kazemi, A, Ni, K, Zhang, J, Chen, PY, Sharma, P, Yu, S, Hu, XS, Niemier, M & Datta, S 2018, 'A ferroelectric field effect transistor based synaptic weight cell', Journal of Physics D: Applied Physics, vol. 51, no. 43, 434001. https://doi.org/10.1088/1361-6463/aad6f8
Jerry, Matthew ; Dutta, Sourav ; Kazemi, Arman ; Ni, Kai ; Zhang, Jianchi ; Chen, Pai Yu ; Sharma, Pankaj ; Yu, Shimeng ; Hu, X. Sharon ; Niemier, Michael ; Datta, Suman. / A ferroelectric field effect transistor based synaptic weight cell. In: Journal of Physics D: Applied Physics. 2018 ; Vol. 51, No. 43.
@article{41fdc66150234cc9a59896e0730f2191,
title = "A ferroelectric field effect transistor based synaptic weight cell",
abstract = "Dense analog synaptic crossbar arrays are a promising candidate for neuromorphic hardware accelerators due to the ability to mitigate data movement by performing in-situ vector-matrix products and weight updates within the storage array itself. However, many analog weight storage cells suffer from long latencies or low dynamic ranges, limiting the achievable performance. In this work, we demonstrate that the voltage-controlled partial polarization switching dynamics in ferroelectric-field-effect transistors (FeFET) can be harnessed to enable a 32 state non-volatile analog synaptic weight cell with large dynamic range (67×) and low latency weight updates (50 ns) for an amplitude modulated pulse scheme.",
keywords = "analog synapse, feld-effect-transistor, ferroelectric, Neuromorphic computing",
author = "Matthew Jerry and Sourav Dutta and Arman Kazemi and Kai Ni and Jianchi Zhang and Chen, {Pai Yu} and Pankaj Sharma and Shimeng Yu and Hu, {X. Sharon} and Michael Niemier and Suman Datta",
year = "2018",
month = "8",
day = "23",
doi = "10.1088/1361-6463/aad6f8",
language = "English (US)",
volume = "51",
journal = "Journal Physics D: Applied Physics",
issn = "0022-3727",
publisher = "IOP Publishing Ltd.",
number = "43",

}

TY - JOUR

T1 - A ferroelectric field effect transistor based synaptic weight cell

AU - Jerry, Matthew

AU - Dutta, Sourav

AU - Kazemi, Arman

AU - Ni, Kai

AU - Zhang, Jianchi

AU - Chen, Pai Yu

AU - Sharma, Pankaj

AU - Yu, Shimeng

AU - Hu, X. Sharon

AU - Niemier, Michael

AU - Datta, Suman

PY - 2018/8/23

Y1 - 2018/8/23

N2 - Dense analog synaptic crossbar arrays are a promising candidate for neuromorphic hardware accelerators due to the ability to mitigate data movement by performing in-situ vector-matrix products and weight updates within the storage array itself. However, many analog weight storage cells suffer from long latencies or low dynamic ranges, limiting the achievable performance. In this work, we demonstrate that the voltage-controlled partial polarization switching dynamics in ferroelectric-field-effect transistors (FeFET) can be harnessed to enable a 32 state non-volatile analog synaptic weight cell with large dynamic range (67×) and low latency weight updates (50 ns) for an amplitude modulated pulse scheme.

AB - Dense analog synaptic crossbar arrays are a promising candidate for neuromorphic hardware accelerators due to the ability to mitigate data movement by performing in-situ vector-matrix products and weight updates within the storage array itself. However, many analog weight storage cells suffer from long latencies or low dynamic ranges, limiting the achievable performance. In this work, we demonstrate that the voltage-controlled partial polarization switching dynamics in ferroelectric-field-effect transistors (FeFET) can be harnessed to enable a 32 state non-volatile analog synaptic weight cell with large dynamic range (67×) and low latency weight updates (50 ns) for an amplitude modulated pulse scheme.

KW - analog synapse

KW - feld-effect-transistor

KW - ferroelectric

KW - Neuromorphic computing

UR - http://www.scopus.com/inward/record.url?scp=85054598499&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054598499&partnerID=8YFLogxK

U2 - 10.1088/1361-6463/aad6f8

DO - 10.1088/1361-6463/aad6f8

M3 - Article

VL - 51

JO - Journal Physics D: Applied Physics

JF - Journal Physics D: Applied Physics

SN - 0022-3727

IS - 43

M1 - 434001

ER -