A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications

Jixuan Wu; Fei Mo; Takuya Saraya; Toshiro Hiramoto; Masaharu Kobayashi

doi:10.1109/VLSITechnology18217.2020.9265062

A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications

Jixuan Wu
, Fei Mo
, Takuya Saraya
, Toshiro Hiramoto
, Masaharu Kobayashi

System Design Research Center (D. Lab
The University of Tokyo

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

28 Citations (Scopus)

Abstract

We have monolithically integrated RRAM array with oxide semiconductor channel access transistor in 3D stack, achieved uniform memory characteristics of 1 T1R cells at each layer, and demonstrated basic functionality of XNOR operation as in-memory computing for binary neural network AI applications, for the first time. The impact of RRAM bit error rate on neural network is also investigated. 3D neural network built by this architecture has high potential to enable area-efficient, low-power and low-latency computing.

Original language	English
Title of host publication	2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728164601
DOIs	https://doi.org/10.1109/VLSITechnology18217.2020.9265062
Publication status	Published - Jun 2020
Externally published	Yes
Event	2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Honolulu, United States Duration: 16 Jun 2020 → 19 Jun 2020

Publication series

Name	Digest of Technical Papers - Symposium on VLSI Technology
Volume	2020-June
ISSN (Print)	0743-1562

Conference

Conference	2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020
Country/Territory	United States
City	Honolulu
Period	16/06/20 → 19/06/20

Access to Document

10.1109/VLSITechnology18217.2020.9265062

Cite this

Wu, J., Mo, F., Saraya, T., Hiramoto, T., & Kobayashi, M. (2020). A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications. In 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings Article 9265062 (Digest of Technical Papers - Symposium on VLSI Technology; Vol. 2020-June). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VLSITechnology18217.2020.9265062

Wu, Jixuan ; Mo, Fei ; Saraya, Takuya et al. / A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications. 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. (Digest of Technical Papers - Symposium on VLSI Technology).

@inproceedings{22c6c656a67246218badeb09e5dadca3,

title = "A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications",

abstract = "We have monolithically integrated RRAM array with oxide semiconductor channel access transistor in 3D stack, achieved uniform memory characteristics of 1 T1R cells at each layer, and demonstrated basic functionality of XNOR operation as in-memory computing for binary neural network AI applications, for the first time. The impact of RRAM bit error rate on neural network is also investigated. 3D neural network built by this architecture has high potential to enable area-efficient, low-power and low-latency computing.",

author = "Jixuan Wu and Fei Mo and Takuya Saraya and Toshiro Hiramoto and Masaharu Kobayashi",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 ; Conference date: 16-06-2020 Through 19-06-2020",

year = "2020",

month = jun,

doi = "10.1109/VLSITechnology18217.2020.9265062",

language = "English",

series = "Digest of Technical Papers - Symposium on VLSI Technology",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings",

address = "United States",

}

Wu, J, Mo, F, Saraya, T, Hiramoto, T & Kobayashi, M 2020, A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications. in 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings., 9265062, Digest of Technical Papers - Symposium on VLSI Technology, vol. 2020-June, Institute of Electrical and Electronics Engineers Inc., 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020, Honolulu, United States, 16/06/20. https://doi.org/10.1109/VLSITechnology18217.2020.9265062

A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications. / Wu, Jixuan; Mo, Fei; Saraya, Takuya et al.
2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. 9265062 (Digest of Technical Papers - Symposium on VLSI Technology; Vol. 2020-June).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications

AU - Wu, Jixuan

AU - Mo, Fei

AU - Saraya, Takuya

AU - Hiramoto, Toshiro

AU - Kobayashi, Masaharu

PY - 2020/6

Y1 - 2020/6

N2 - We have monolithically integrated RRAM array with oxide semiconductor channel access transistor in 3D stack, achieved uniform memory characteristics of 1 T1R cells at each layer, and demonstrated basic functionality of XNOR operation as in-memory computing for binary neural network AI applications, for the first time. The impact of RRAM bit error rate on neural network is also investigated. 3D neural network built by this architecture has high potential to enable area-efficient, low-power and low-latency computing.

AB - We have monolithically integrated RRAM array with oxide semiconductor channel access transistor in 3D stack, achieved uniform memory characteristics of 1 T1R cells at each layer, and demonstrated basic functionality of XNOR operation as in-memory computing for binary neural network AI applications, for the first time. The impact of RRAM bit error rate on neural network is also investigated. 3D neural network built by this architecture has high potential to enable area-efficient, low-power and low-latency computing.

UR - https://www.scopus.com/pages/publications/85098189319

U2 - 10.1109/VLSITechnology18217.2020.9265062

DO - 10.1109/VLSITechnology18217.2020.9265062

M3 - Conference contribution

AN - SCOPUS:85098189319

T3 - Digest of Technical Papers - Symposium on VLSI Technology

BT - 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020

Y2 - 16 June 2020 through 19 June 2020

ER -

Wu J, Mo F, Saraya T, Hiramoto T, Kobayashi M. A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications. In 2020 IEEE Symposium on VLSI Technology, VLSI Technology 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2020. 9265062. (Digest of Technical Papers - Symposium on VLSI Technology). doi: 10.1109/VLSITechnology18217.2020.9265062

A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-memory Computing in Quantized Neural Network AI Applications

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this