Supervised learning in spiking neural networks with MLC PCM synapses

S. R. Nandakumar; I. Boybat; M. Le Gallo; A. Sebastian; B. Rajendran; E. Eleftheriou

doi:10.1109/DRC.2017.7999481

Supervised learning in spiking neural networks with MLC PCM synapses

S. R. Nandakumar, I. Boybat, M. Le Gallo, A. Sebastian, B. Rajendran, E. Eleftheriou

Electrical and Computer Engineering

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

11 Scopus citations

Abstract

We demonstrate for the first time, the feasibility of supervised learning in third generation Spiking Neural Networks (SNNs) using multi-level cell (MLC) phase change memory (PCM) synapses [1]. We highlight two key novel contributions: (i) As opposed to second generation neural networks that are used in machine learning algorithms [2], or spike timing dependent plasticity based unsupervised learning in SNNs [3], we use a spike-triggered supervised learning algorithm (NormAD [4]) for the weight updates. (ii) SNN learning capability is demonstrated using a comprehensive phenomenological model of MLC PCM that accurately captures the statistics of programming inter-cell and intra-cell variability. This work is a harbinger to efficient supervised SNN learning systems.

Original language	English (US)
Title of host publication	75th Annual Device Research Conference, DRC 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781509063277
DOIs	https://doi.org/10.1109/DRC.2017.7999481
State	Published - Aug 1 2017
Event	75th Annual Device Research Conference, DRC 2017 - South Bend, United States Duration: Jun 25 2017 → Jun 28 2017

Publication series

Name	Device Research Conference - Conference Digest, DRC
ISSN (Print)	1548-3770

Other

Other	75th Annual Device Research Conference, DRC 2017
Country/Territory	United States
City	South Bend
Period	6/25/17 → 6/28/17

All Science Journal Classification (ASJC) codes

Electrical and Electronic Engineering

Access to Document

10.1109/DRC.2017.7999481

Cite this

Nandakumar, S. R., Boybat, I., Le Gallo, M., Sebastian, A., Rajendran, B., & Eleftheriou, E. (2017). Supervised learning in spiking neural networks with MLC PCM synapses. In 75th Annual Device Research Conference, DRC 2017 Article 7999481 (Device Research Conference - Conference Digest, DRC). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DRC.2017.7999481

@inproceedings{8339d2d4bb5841329722b58b59c8a15b,

title = "Supervised learning in spiking neural networks with MLC PCM synapses",

abstract = "We demonstrate for the first time, the feasibility of supervised learning in third generation Spiking Neural Networks (SNNs) using multi-level cell (MLC) phase change memory (PCM) synapses [1]. We highlight two key novel contributions: (i) As opposed to second generation neural networks that are used in machine learning algorithms [2], or spike timing dependent plasticity based unsupervised learning in SNNs [3], we use a spike-triggered supervised learning algorithm (NormAD [4]) for the weight updates. (ii) SNN learning capability is demonstrated using a comprehensive phenomenological model of MLC PCM that accurately captures the statistics of programming inter-cell and intra-cell variability. This work is a harbinger to efficient supervised SNN learning systems.",

author = "Nandakumar, {S. R.} and I. Boybat and {Le Gallo}, M. and A. Sebastian and B. Rajendran and E. Eleftheriou",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 75th Annual Device Research Conference, DRC 2017 ; Conference date: 25-06-2017 Through 28-06-2017",

year = "2017",

month = aug,

day = "1",

doi = "10.1109/DRC.2017.7999481",

language = "English (US)",

series = "Device Research Conference - Conference Digest, DRC",

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

booktitle = "75th Annual Device Research Conference, DRC 2017",

address = "United States",

}

Nandakumar, SR, Boybat, I, Le Gallo, M, Sebastian, A, Rajendran, B & Eleftheriou, E 2017, Supervised learning in spiking neural networks with MLC PCM synapses. in 75th Annual Device Research Conference, DRC 2017., 7999481, Device Research Conference - Conference Digest, DRC, Institute of Electrical and Electronics Engineers Inc., 75th Annual Device Research Conference, DRC 2017, South Bend, United States, 6/25/17. https://doi.org/10.1109/DRC.2017.7999481

Supervised learning in spiking neural networks with MLC PCM synapses. / Nandakumar, S. R.; Boybat, I.; Le Gallo, M. et al.
75th Annual Device Research Conference, DRC 2017. Institute of Electrical and Electronics Engineers Inc., 2017. 7999481 (Device Research Conference - Conference Digest, DRC).

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

TY - GEN

T1 - Supervised learning in spiking neural networks with MLC PCM synapses

AU - Nandakumar, S. R.

AU - Boybat, I.

AU - Le Gallo, M.

AU - Sebastian, A.

AU - Rajendran, B.

AU - Eleftheriou, E.

PY - 2017/8/1

Y1 - 2017/8/1

N2 - We demonstrate for the first time, the feasibility of supervised learning in third generation Spiking Neural Networks (SNNs) using multi-level cell (MLC) phase change memory (PCM) synapses [1]. We highlight two key novel contributions: (i) As opposed to second generation neural networks that are used in machine learning algorithms [2], or spike timing dependent plasticity based unsupervised learning in SNNs [3], we use a spike-triggered supervised learning algorithm (NormAD [4]) for the weight updates. (ii) SNN learning capability is demonstrated using a comprehensive phenomenological model of MLC PCM that accurately captures the statistics of programming inter-cell and intra-cell variability. This work is a harbinger to efficient supervised SNN learning systems.

AB - We demonstrate for the first time, the feasibility of supervised learning in third generation Spiking Neural Networks (SNNs) using multi-level cell (MLC) phase change memory (PCM) synapses [1]. We highlight two key novel contributions: (i) As opposed to second generation neural networks that are used in machine learning algorithms [2], or spike timing dependent plasticity based unsupervised learning in SNNs [3], we use a spike-triggered supervised learning algorithm (NormAD [4]) for the weight updates. (ii) SNN learning capability is demonstrated using a comprehensive phenomenological model of MLC PCM that accurately captures the statistics of programming inter-cell and intra-cell variability. This work is a harbinger to efficient supervised SNN learning systems.

UR - http://www.scopus.com/inward/record.url?scp=85028032578&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028032578&partnerID=8YFLogxK

U2 - 10.1109/DRC.2017.7999481

DO - 10.1109/DRC.2017.7999481

M3 - Conference contribution

AN - SCOPUS:85028032578

T3 - Device Research Conference - Conference Digest, DRC

BT - 75th Annual Device Research Conference, DRC 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 75th Annual Device Research Conference, DRC 2017

Y2 - 25 June 2017 through 28 June 2017

ER -

Supervised learning in spiking neural networks with MLC PCM synapses

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this