Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays

Shruti R. Kulkarni; Shihui Yin; Jae Sun Seo; Bipin Rajendran

doi:10.1109/ICEE56203.2022.10118061

Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays

Shruti R. Kulkarni, Shihui Yin, Jae Sun Seo, Bipin Rajendran

Learning Systems

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

Abstract

In this paper, we present a scalable digital hardware accelerator based on non-volatile memory arrays capable of realizing deep convolutional spiking neural networks (SNNs). Our design studies are conducted using a compact model for spin-transfer torque random access memory (STT-RAM) devices. Large networks are realized by tiling multiple cores which communicate by transmitting spike packets via an on-chip routing network. Compared to an equivalent SRAM based core design, we show that the STT-RAM based design achieves nearly 15X higher GSOPS (Synaptic Operations per Second) per Watt per mm2 making it a promising platform for realizing systems with significant area and power limitations.

Original language	English
Title of host publication	2022 IEEE International Conference on Emerging Electronics, ICEE 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665491853
DOIs	https://doi.org/10.1109/ICEE56203.2022.10118061
State	Published - 2022
Event	2022 IEEE International Conference on Emerging Electronics, ICEE 2022 - Bangalore, India Duration: Dec 11 2022 → Dec 14 2022

Publication series

Name	2022 IEEE International Conference on Emerging Electronics, ICEE 2022

Conference

Conference	2022 IEEE International Conference on Emerging Electronics, ICEE 2022
Country/Territory	India
City	Bangalore
Period	12/11/22 → 12/14/22

Funding

The authors acknowledge the valuable discussions with Anakha V. Babu from NJIT and Shreyas K. Venkataramana-iah from ASU. This research was supported in part by the Semiconductor Research Corporation (2016-SD-2717). Notice: This manuscript has been authored in part by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a non-exclusive, paid up, irrevocable, world-wide license to publish or reproduce the published form of the manuscript, or allow others to do so, for U.S. Government purposes. The DOE will provide public access to these results in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

Keywords

Spiking neural networks
Spin Transfer Torque RAM
neuromorphic accelerators
non-volatile memory

Access to Document

10.1109/ICEE56203.2022.10118061

Cite this

Kulkarni, S. R., Yin, S., Seo, J. S., & Rajendran, B. (2022). Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays. In 2022 IEEE International Conference on Emerging Electronics, ICEE 2022 (2022 IEEE International Conference on Emerging Electronics, ICEE 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICEE56203.2022.10118061

@inproceedings{d1051d3075ac402d81aeefe3a687d0f3,

title = "Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays",

abstract = "In this paper, we present a scalable digital hardware accelerator based on non-volatile memory arrays capable of realizing deep convolutional spiking neural networks (SNNs). Our design studies are conducted using a compact model for spin-transfer torque random access memory (STT-RAM) devices. Large networks are realized by tiling multiple cores which communicate by transmitting spike packets via an on-chip routing network. Compared to an equivalent SRAM based core design, we show that the STT-RAM based design achieves nearly 15X higher GSOPS (Synaptic Operations per Second) per Watt per mm2 making it a promising platform for realizing systems with significant area and power limitations.",

keywords = "Spiking neural networks, Spin Transfer Torque RAM, neuromorphic accelerators, non-volatile memory",

author = "Kulkarni, {Shruti R.} and Shihui Yin and Seo, {Jae Sun} and Bipin Rajendran",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Conference on Emerging Electronics, ICEE 2022 ; Conference date: 11-12-2022 Through 14-12-2022",

year = "2022",

doi = "10.1109/ICEE56203.2022.10118061",

language = "English",

series = "2022 IEEE International Conference on Emerging Electronics, ICEE 2022",

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

booktitle = "2022 IEEE International Conference on Emerging Electronics, ICEE 2022",

}

Kulkarni, SR, Yin, S, Seo, JS & Rajendran, B 2022, Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays. in 2022 IEEE International Conference on Emerging Electronics, ICEE 2022. 2022 IEEE International Conference on Emerging Electronics, ICEE 2022, Institute of Electrical and Electronics Engineers Inc., 2022 IEEE International Conference on Emerging Electronics, ICEE 2022, Bangalore, India, 12/11/22. https://doi.org/10.1109/ICEE56203.2022.10118061

Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays. / Kulkarni, Shruti R.; Yin, Shihui; Seo, Jae Sun et al.
2022 IEEE International Conference on Emerging Electronics, ICEE 2022. Institute of Electrical and Electronics Engineers Inc., 2022. (2022 IEEE International Conference on Emerging Electronics, ICEE 2022).

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

TY - GEN

T1 - Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays

AU - Kulkarni, Shruti R.

AU - Yin, Shihui

AU - Seo, Jae Sun

AU - Rajendran, Bipin

PY - 2022

Y1 - 2022

N2 - In this paper, we present a scalable digital hardware accelerator based on non-volatile memory arrays capable of realizing deep convolutional spiking neural networks (SNNs). Our design studies are conducted using a compact model for spin-transfer torque random access memory (STT-RAM) devices. Large networks are realized by tiling multiple cores which communicate by transmitting spike packets via an on-chip routing network. Compared to an equivalent SRAM based core design, we show that the STT-RAM based design achieves nearly 15X higher GSOPS (Synaptic Operations per Second) per Watt per mm2 making it a promising platform for realizing systems with significant area and power limitations.

AB - In this paper, we present a scalable digital hardware accelerator based on non-volatile memory arrays capable of realizing deep convolutional spiking neural networks (SNNs). Our design studies are conducted using a compact model for spin-transfer torque random access memory (STT-RAM) devices. Large networks are realized by tiling multiple cores which communicate by transmitting spike packets via an on-chip routing network. Compared to an equivalent SRAM based core design, we show that the STT-RAM based design achieves nearly 15X higher GSOPS (Synaptic Operations per Second) per Watt per mm2 making it a promising platform for realizing systems with significant area and power limitations.

KW - Spiking neural networks

KW - Spin Transfer Torque RAM

KW - neuromorphic accelerators

KW - non-volatile memory

UR - http://www.scopus.com/inward/record.url?scp=85160536306&partnerID=8YFLogxK

U2 - 10.1109/ICEE56203.2022.10118061

DO - 10.1109/ICEE56203.2022.10118061

M3 - Conference contribution

AN - SCOPUS:85160536306

T3 - 2022 IEEE International Conference on Emerging Electronics, ICEE 2022

BT - 2022 IEEE International Conference on Emerging Electronics, ICEE 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 IEEE International Conference on Emerging Electronics, ICEE 2022

Y2 - 11 December 2022 through 14 December 2022

ER -

Neuromorphic Accelerator for Deep Spiking Neural Networks with NVM Crossbar Arrays

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this