Architectural design exploration for neuromorphic processors with memristive synapses

Qian Wang; Yongtae Kim; Peng Li

doi:10.1109/NANO.2014.6967962

Architectural design exploration for neuromorphic processors with memristive synapses

Qian Wang, Yongtae Kim, Peng Li

Texas A&M University

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

9 Scopus citations

Abstract

Due to their nonvolatile nature, excellent scalability and high density, memristive nanodevices provide a promising solution for low-cost on-chip storage. Integrating memristor-based synaptic crossbars into digital neuromorphic processors (DNPs) may facilitate efficient realization of brain inspired computing. This paper investigates architectural design exploration of DNPs with memristive synapses by proposing two synapse readout schemes. The key design tradeoffs involving different analog-to-digital conversions and memory accessing styles are thoroughly investigated.

Original language	English
Title of host publication	Proceedings of the IEEE Conference on Nanotechnology
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	962-966
Number of pages	5
ISBN (Electronic)	9781479956227
DOIs	https://doi.org/10.1109/NANO.2014.6967962
State	Published - 26 Nov 2014
Event	2014 14th IEEE International Conference on Nanotechnology, IEEE-NANO 2014 - Toronto, Canada Duration: 18 Aug 2014 → 21 Aug 2014

Publication series

Name	Proceedings of the IEEE Conference on Nanotechnology
ISSN (Electronic)	1944-9399

Conference

Conference	2014 14th IEEE International Conference on Nanotechnology, IEEE-NANO 2014
Country/Territory	Canada
City	Toronto
Period	18/08/14 → 21/08/14

Access to Document

10.1109/NANO.2014.6967962

Cite this

@inproceedings{8016e19efb2543aa93a6aa137efb359b,

title = "Architectural design exploration for neuromorphic processors with memristive synapses",

abstract = "Due to their nonvolatile nature, excellent scalability and high density, memristive nanodevices provide a promising solution for low-cost on-chip storage. Integrating memristor-based synaptic crossbars into digital neuromorphic processors (DNPs) may facilitate efficient realization of brain inspired computing. This paper investigates architectural design exploration of DNPs with memristive synapses by proposing two synapse readout schemes. The key design tradeoffs involving different analog-to-digital conversions and memory accessing styles are thoroughly investigated.",

author = "Qian Wang and Yongtae Kim and Peng Li",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 14th IEEE International Conference on Nanotechnology, IEEE-NANO 2014 ; Conference date: 18-08-2014 Through 21-08-2014",

year = "2014",

month = nov,

day = "26",

doi = "10.1109/NANO.2014.6967962",

language = "English",

series = "Proceedings of the IEEE Conference on Nanotechnology",

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

pages = "962--966",

booktitle = "Proceedings of the IEEE Conference on Nanotechnology",

address = "United States",

}

Wang, Q, Kim, Y & Li, P 2014, Architectural design exploration for neuromorphic processors with memristive synapses. in Proceedings of the IEEE Conference on Nanotechnology., 6967962, Proceedings of the IEEE Conference on Nanotechnology, Institute of Electrical and Electronics Engineers Inc., pp. 962-966, 2014 14th IEEE International Conference on Nanotechnology, IEEE-NANO 2014, Toronto, Canada, 18/08/14. https://doi.org/10.1109/NANO.2014.6967962

Architectural design exploration for neuromorphic processors with memristive synapses. / Wang, Qian; Kim, Yongtae; Li, Peng.
Proceedings of the IEEE Conference on Nanotechnology. Institute of Electrical and Electronics Engineers Inc., 2014. p. 962-966 6967962 (Proceedings of the IEEE Conference on Nanotechnology).

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

TY - GEN

T1 - Architectural design exploration for neuromorphic processors with memristive synapses

AU - Wang, Qian

AU - Kim, Yongtae

AU - Li, Peng

PY - 2014/11/26

Y1 - 2014/11/26

N2 - Due to their nonvolatile nature, excellent scalability and high density, memristive nanodevices provide a promising solution for low-cost on-chip storage. Integrating memristor-based synaptic crossbars into digital neuromorphic processors (DNPs) may facilitate efficient realization of brain inspired computing. This paper investigates architectural design exploration of DNPs with memristive synapses by proposing two synapse readout schemes. The key design tradeoffs involving different analog-to-digital conversions and memory accessing styles are thoroughly investigated.

AB - Due to their nonvolatile nature, excellent scalability and high density, memristive nanodevices provide a promising solution for low-cost on-chip storage. Integrating memristor-based synaptic crossbars into digital neuromorphic processors (DNPs) may facilitate efficient realization of brain inspired computing. This paper investigates architectural design exploration of DNPs with memristive synapses by proposing two synapse readout schemes. The key design tradeoffs involving different analog-to-digital conversions and memory accessing styles are thoroughly investigated.

UR - http://www.scopus.com/inward/record.url?scp=84919465728&partnerID=8YFLogxK

U2 - 10.1109/NANO.2014.6967962

DO - 10.1109/NANO.2014.6967962

M3 - Conference contribution

AN - SCOPUS:84919465728

T3 - Proceedings of the IEEE Conference on Nanotechnology

SP - 962

EP - 966

BT - Proceedings of the IEEE Conference on Nanotechnology

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 14th IEEE International Conference on Nanotechnology, IEEE-NANO 2014

Y2 - 18 August 2014 through 21 August 2014

ER -

Architectural design exploration for neuromorphic processors with memristive synapses

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this