Oscillatory Neural Networks for Edge AI Computing

Corentin Delacour; Stefania Carapezzi; Madeleine Abernot; Gabriele Boschetto; Nadine Azémard; Jérémie Salles; Thierry Gil; Aida Todri-Sanial

doi:10.1109/ISVLSI51109.2021.00066

Oscillatory Neural Networks for Edge AI Computing

Corentin Delacour
, Stefania Carapezzi
, Madeleine Abernot
, Gabriele Boschetto
, Nadine Azémard
, Jérémie Salles
, Thierry Gil
, Aida Todri-Sanial

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

28 Citations (Scopus)

Abstract

In this paper, we showcase the innovative concept of implementing Oscillatory Neural Networks (ONNs) for neuromorphic computing with beyond-CMOS devices based on vanadium dioxide to mimic neurons and resistors to emulate synapses. We explore ONN technology potentials from device to analog circuit-level simulations. We report that ONN behaves like an associative memory and can implement energy-based models such as Hopfield Neural Networks on edge devices. Finally, as a proof of concept, a reconfigurable digital ONN is implemented on FPGA for pattern recognition tasks.

Original language	English
Title of host publication	2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)
Publisher	Institute of Electrical and Electronics Engineers
Pages	326-331
Number of pages	6
ISBN (Electronic)	978-1-6654-3946-6
DOIs	https://doi.org/10.1109/ISVLSI51109.2021.00066
Publication status	Published - Jul 2021
Externally published	Yes
Event	2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI 2021) - Tampa, United States Duration: 7 Jul 2021 → 9 Jul 2021

Conference

Conference	2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI 2021)
Country/Territory	United States
City	Tampa
Period	7/07/21 → 9/07/21

Keywords

Oscillatory neural networks (ONN)
Internet-of-Things (IoT)
Edge Artificial Intelligence (edge AI)
Neuromorphic Computing
Beyond-CMOS devices
Technology Computer-Aided Design (TCAD)
FPGA

Access to Document

10.1109/ISVLSI51109.2021.00066

Cite this

@inproceedings{d0f98b142b3e4826b772396fc7a430f4,

title = "Oscillatory Neural Networks for Edge AI Computing",

abstract = "In this paper, we showcase the innovative concept of implementing Oscillatory Neural Networks (ONNs) for neuromorphic computing with beyond-CMOS devices based on vanadium dioxide to mimic neurons and resistors to emulate synapses. We explore ONN technology potentials from device to analog circuit-level simulations. We report that ONN behaves like an associative memory and can implement energy-based models such as Hopfield Neural Networks on edge devices. Finally, as a proof of concept, a reconfigurable digital ONN is implemented on FPGA for pattern recognition tasks.",

keywords = "Oscillatory neural networks (ONN), Internet-of-Things (IoT), Edge Artificial Intelligence (edge AI), Neuromorphic Computing, Beyond-CMOS devices, Technology Computer-Aided Design (TCAD), FPGA",

author = "Corentin Delacour and Stefania Carapezzi and Madeleine Abernot and Gabriele Boschetto and Nadine Az{\'e}mard and J{\'e}r{\'e}mie Salles and Thierry Gil and Aida Todri-Sanial",

year = "2021",

month = jul,

doi = "10.1109/ISVLSI51109.2021.00066",

language = "English",

pages = "326--331",

booktitle = "2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI 2021) ; Conference date: 07-07-2021 Through 09-07-2021",

}

Delacour, C, Carapezzi, S, Abernot, M, Boschetto, G, Azémard, N, Salles, J, Gil, T & Todri-Sanial, A 2021, Oscillatory Neural Networks for Edge AI Computing. in 2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI). Institute of Electrical and Electronics Engineers, pp. 326-331, 2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI 2021), Tampa, Florida, United States, 7/07/21. https://doi.org/10.1109/ISVLSI51109.2021.00066

TY - GEN

T1 - Oscillatory Neural Networks for Edge AI Computing

AU - Delacour, Corentin

AU - Carapezzi, Stefania

AU - Abernot, Madeleine

AU - Boschetto, Gabriele

AU - Azémard, Nadine

AU - Salles, Jérémie

AU - Gil, Thierry

AU - Todri-Sanial, Aida

PY - 2021/7

Y1 - 2021/7

N2 - In this paper, we showcase the innovative concept of implementing Oscillatory Neural Networks (ONNs) for neuromorphic computing with beyond-CMOS devices based on vanadium dioxide to mimic neurons and resistors to emulate synapses. We explore ONN technology potentials from device to analog circuit-level simulations. We report that ONN behaves like an associative memory and can implement energy-based models such as Hopfield Neural Networks on edge devices. Finally, as a proof of concept, a reconfigurable digital ONN is implemented on FPGA for pattern recognition tasks.

AB - In this paper, we showcase the innovative concept of implementing Oscillatory Neural Networks (ONNs) for neuromorphic computing with beyond-CMOS devices based on vanadium dioxide to mimic neurons and resistors to emulate synapses. We explore ONN technology potentials from device to analog circuit-level simulations. We report that ONN behaves like an associative memory and can implement energy-based models such as Hopfield Neural Networks on edge devices. Finally, as a proof of concept, a reconfigurable digital ONN is implemented on FPGA for pattern recognition tasks.

KW - Oscillatory neural networks (ONN)

KW - Internet-of-Things (IoT)

KW - Edge Artificial Intelligence (edge AI)

KW - Neuromorphic Computing

KW - Beyond-CMOS devices

KW - Technology Computer-Aided Design (TCAD)

KW - FPGA

UR - https://www.scopus.com/pages/publications/85114963563

U2 - 10.1109/ISVLSI51109.2021.00066

DO - 10.1109/ISVLSI51109.2021.00066

M3 - Conference contribution

SP - 326

EP - 331

BT - 2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)

PB - Institute of Electrical and Electronics Engineers

T2 - 2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI 2021)

Y2 - 7 July 2021 through 9 July 2021

ER -

Oscillatory Neural Networks for Edge AI Computing

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this