Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > cs > arXiv:1912.12636

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Computer Science > Emerging Technologies

arXiv:1912.12636 (cs)
[Submitted on 29 Dec 2019 (v1), last revised 29 May 2022 (this version, v2)]

Title:Training of Quantized Deep Neural Networks using a Magnetic Tunnel Junction-Based Synapse

Authors:Tzofnat Greenberg Toledo, Ben Perach, Itay Hubara, Daniel Soudry, Shahar Kvatinsky
View a PDF of the paper titled Training of Quantized Deep Neural Networks using a Magnetic Tunnel Junction-Based Synapse, by Tzofnat Greenberg Toledo and 3 other authors
View PDF
Abstract:Quantized neural networks (QNNs) are being actively researched as a solution for the computational complexity and memory intensity of deep neural networks. This has sparked efforts to develop algorithms that support both inference and training with quantized weight and activation values, without sacrificing accuracy. A recent example is the GXNOR framework for stochastic training of ternary (TNN) and binary (BNN) neural networks. In this paper, we show how magnetic tunnel junction (MTJ) devices can be used to support QNN training. We introduce a novel hardware synapse circuit that uses the MTJ stochastic behavior to support the quantize update. The proposed circuit enables processing near memory (PNM) of QNN training, which subsequently reduces data movement. We simulated MTJ-based stochastic training of a TNN over the MNIST, SVHN, and CIFAR10 datasets and achieved an accuracy of 98.61%, 93.99% and 82.71%, respectively (less than 1% degradation compared to the GXNOR algorithm). We evaluated the synapse array performance potential and showed that the proposed synapse circuit can train ternary networks in situ, with 18.3TOPs/W for feedforward and 3TOPs/W for weight update.
Comments: Published in Semiconductor Science and Technology, Vol 36
Subjects: Emerging Technologies (cs.ET); Hardware Architecture (cs.AR); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
Cite as: arXiv:1912.12636 [cs.ET]
  (or arXiv:1912.12636v2 [cs.ET] for this version)
  https://doi.org/10.48550/arXiv.1912.12636
arXiv-issued DOI via DataCite
Journal reference: Semicond. Sci. Technol. 36 114003 (2021)
Related DOI: https://doi.org/10.1088/1361-6641/ac251b
DOI(s) linking to related resources

Submission history

From: Tzofnat Greenberg-Toledo [view email]
[v1] Sun, 29 Dec 2019 11:36:32 UTC (1,504 KB)
[v2] Sun, 29 May 2022 07:16:41 UTC (2,731 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Training of Quantized Deep Neural Networks using a Magnetic Tunnel Junction-Based Synapse, by Tzofnat Greenberg Toledo and 3 other authors
  • View PDF
  • TeX Source
  • Other Formats
license icon view license
Current browse context:
cs.ET
< prev   |   next >
new | recent | 2019-12
Change to browse by:
cs
cs.AR
cs.LG
cs.NE

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar

DBLP - CS Bibliography

listing | bibtex
Daniel Soudry
Shahar Kvatinsky
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack