Energy Efficient Neuromorphic Computing with beyond-CMOS Oscillatory Neural Networks

Corentin Delacour; Stefania Carapezzi; Gabriele Boschetto; Aida Todri-Sanial

Energy Efficient Neuromorphic Computing with beyond-CMOS Oscillatory Neural Networks

Acte de colloque

Open Access

Energy Efficient Neuromorphic Computing with beyond-CMOS Oscillatory Neural Networks

Corentin Delacour, Stefania Carapezzi, Gabriele Boschetto et Aida Todri-Sanial

ICONS 2021 - International Conference on Neuromorphic Systems (Oak Ridge (Virtual), United States, 27/07/2021–29/07/2021)

Résumé

ONN

Neuromorphic circuit

Beyond-CMOS

VO2

MoS2 Memristor

Oscillatory Neural Networks (ONNs) are non-von Neumann architectures where information is encoded in phase relations between coupled oscillators. In this work, we present the concept of ONN based on beyond-CMOS devices to reduce the energy footprint of neuromorphic circuits. We investigate oscillating neurons made of vanadium dioxide material (VO2) and synapses based on molybdenum disulfide (MoS2) memristors to emulate synaptic plasticity.

Fichiers et liens (1)

url

Find in HALAfficher

Indicateurs

1 Consultations de la notice

Détails

Titre: Energy Efficient Neuromorphic Computing with beyond-CMOS Oscillatory Neural Networks
Créateurs - sans rôle: Corentin Delacour - Smart Integrated Electronic Systems
Stefania Carapezzi - Smart Integrated Electronic Systems
Gabriele Boschetto - Smart Integrated Electronic Systems
Aida Todri-Sanial - Smart Integrated Electronic Systems
Colloque: ICONS 2021 - International Conference on Neuromorphic Systems (Oak Ridge (Virtual), United States, 27/07/2021–29/07/2021)
Note de subvention: This work has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 871501.
Identifiants: 9947663809311
Unité académique: Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier - LIRMM
Langue: English
Type de ressource: Conference proceeding
Champs locaux: lirmm-03229262