Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning

Deepak Kadetotad; Zihan Xu; Abinash Mohanty; Pai Yu Chen; Binbin Lin; Jieping Ye; Sarma Vrudhula; Shimeng Yu; Yu Cao; Jae-sun Seo

doi:10.1109/BioCAS.2014.6981781

Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning

Deepak Kadetotad, Zihan Xu, Abinash Mohanty, Pai Yu Chen, Binbin Lin, Jieping Ye, Sarma Vrudhula, Shimeng Yu, Yu Cao, Jae-sun Seo

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

10 Scopus citations

Abstract

This paper proposes a parallel architecture with resistive crosspoint array. The design of its two essential operations, Read and Write, is inspired by the biophysical behavior of a neural system, such as integrate-and-fire and time-dependent synaptic plasticity. The proposed hardware consists of an array with resistive random access memory (RRAM) and CMOS peripheral circuits, which perform matrix product and dictionary update in a fully parallel fashion, at the speed that is independent of the matrix dimension. The entire system is implemented in 65nm CMOS technology with RRAM to realize high-speed unsupervised dictionary learning. As compared to state-of-the-art software approach, it achieves more than 3000X speedup, enabling real-time feature extraction on a single chip.

Original language	English (US)
Title of host publication	IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	536-539
Number of pages	4
ISBN (Electronic)	9781479923465
DOIs	https://doi.org/10.1109/BioCAS.2014.6981781
State	Published - Dec 9 2014
Event	10th IEEE Biomedical Circuits and Systems Conference, BioCAS 2014 - Lausanne, Switzerland Duration: Oct 22 2014 → Oct 24 2014

Publication series

Name	IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings

Other

Other	10th IEEE Biomedical Circuits and Systems Conference, BioCAS 2014
Country/Territory	Switzerland
City	Lausanne
Period	10/22/14 → 10/24/14

ASJC Scopus subject areas

Hardware and Architecture
Biomedical Engineering

Access to Document

10.1109/BioCAS.2014.6981781

Cite this

Kadetotad, D., Xu, Z., Mohanty, A., Chen, P. Y., Lin, B., Ye, J., Vrudhula, S., Yu, S., Cao, Y., & Seo, J. (2014). Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning. In IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings (pp. 536-539). Article 6981781 (IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BioCAS.2014.6981781

Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning. / Kadetotad, Deepak; Xu, Zihan; Mohanty, Abinash et al.
IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2014. p. 536-539 6981781 (IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings).

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

Kadetotad, D, Xu, Z, Mohanty, A, Chen, PY, Lin, B, Ye, J, Vrudhula, S, Yu, S, Cao, Y & Seo, J 2014, Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning. in IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings., 6981781, IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 536-539, 10th IEEE Biomedical Circuits and Systems Conference, BioCAS 2014, Lausanne, Switzerland, 10/22/14. https://doi.org/10.1109/BioCAS.2014.6981781

Kadetotad D, Xu Z, Mohanty A, Chen PY, Lin B, Ye J et al. Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning. In IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2014. p. 536-539. 6981781. (IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings). doi: 10.1109/BioCAS.2014.6981781

Kadetotad, Deepak ; Xu, Zihan ; Mohanty, Abinash et al. / Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning. IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 536-539 (IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings).

@inproceedings{b6518a5bf8ed4c26b013169d1ec32692,

title = "Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning",

abstract = "This paper proposes a parallel architecture with resistive crosspoint array. The design of its two essential operations, Read and Write, is inspired by the biophysical behavior of a neural system, such as integrate-and-fire and time-dependent synaptic plasticity. The proposed hardware consists of an array with resistive random access memory (RRAM) and CMOS peripheral circuits, which perform matrix product and dictionary update in a fully parallel fashion, at the speed that is independent of the matrix dimension. The entire system is implemented in 65nm CMOS technology with RRAM to realize high-speed unsupervised dictionary learning. As compared to state-of-the-art software approach, it achieves more than 3000X speedup, enabling real-time feature extraction on a single chip.",

author = "Deepak Kadetotad and Zihan Xu and Abinash Mohanty and Chen, {Pai Yu} and Binbin Lin and Jieping Ye and Sarma Vrudhula and Shimeng Yu and Yu Cao and Jae-sun Seo",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 10th IEEE Biomedical Circuits and Systems Conference, BioCAS 2014 ; Conference date: 22-10-2014 Through 24-10-2014",

year = "2014",

month = dec,

day = "9",

doi = "10.1109/BioCAS.2014.6981781",

language = "English (US)",

series = "IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings",

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

pages = "536--539",

booktitle = "IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings",

}

TY - GEN

T1 - Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning

AU - Kadetotad, Deepak

AU - Xu, Zihan

AU - Mohanty, Abinash

AU - Chen, Pai Yu

AU - Lin, Binbin

AU - Ye, Jieping

AU - Vrudhula, Sarma

AU - Yu, Shimeng

AU - Cao, Yu

AU - Seo, Jae-sun

PY - 2014/12/9

Y1 - 2014/12/9

N2 - This paper proposes a parallel architecture with resistive crosspoint array. The design of its two essential operations, Read and Write, is inspired by the biophysical behavior of a neural system, such as integrate-and-fire and time-dependent synaptic plasticity. The proposed hardware consists of an array with resistive random access memory (RRAM) and CMOS peripheral circuits, which perform matrix product and dictionary update in a fully parallel fashion, at the speed that is independent of the matrix dimension. The entire system is implemented in 65nm CMOS technology with RRAM to realize high-speed unsupervised dictionary learning. As compared to state-of-the-art software approach, it achieves more than 3000X speedup, enabling real-time feature extraction on a single chip.

AB - This paper proposes a parallel architecture with resistive crosspoint array. The design of its two essential operations, Read and Write, is inspired by the biophysical behavior of a neural system, such as integrate-and-fire and time-dependent synaptic plasticity. The proposed hardware consists of an array with resistive random access memory (RRAM) and CMOS peripheral circuits, which perform matrix product and dictionary update in a fully parallel fashion, at the speed that is independent of the matrix dimension. The entire system is implemented in 65nm CMOS technology with RRAM to realize high-speed unsupervised dictionary learning. As compared to state-of-the-art software approach, it achieves more than 3000X speedup, enabling real-time feature extraction on a single chip.

UR - http://www.scopus.com/inward/record.url?scp=84920545941&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920545941&partnerID=8YFLogxK

U2 - 10.1109/BioCAS.2014.6981781

DO - 10.1109/BioCAS.2014.6981781

M3 - Conference contribution

AN - SCOPUS:84920545941

T3 - IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings

SP - 536

EP - 539

BT - IEEE 2014 Biomedical Circuits and Systems Conference, BioCAS 2014 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th IEEE Biomedical Circuits and Systems Conference, BioCAS 2014

Y2 - 22 October 2014 through 24 October 2014

ER -

Neurophysics-inspired parallel architecture with resistive crosspoint array for dictionary learning

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this