Titre :

Exploiting Non-idealities of Resistive Switching Memories for Efficient Machine Learning

Auteur(s) :

Yon, Victor [Auteur]
Institut Interdisciplinaire d'Innovation Technologique [Sherbrooke] [3IT]
Laboratoire Nanotechnologies et Nanosystèmes [Sherbrooke] [LN2]
Amirsoleimani, Amirali [Auteur]
York University [Toronto]
Alibart, Fabien [Auteur]
Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520 [IEMN]
Nanostructures, nanoComponents & Molecules - IEMN [NCM - IEMN]
Laboratoire Nanotechnologies et Nanosystèmes [Sherbrooke] [LN2]
Melko, Roger [Auteur]
Department of Physics and Astronomy [Waterloo]
Perimeter Institute for Theoretical Physics [Waterloo]
Drouin, Dominique [Auteur]
Institut Interdisciplinaire d'Innovation Technologique [Sherbrooke] [3IT]
Laboratoire Nanotechnologies et Nanosystèmes [Sherbrooke] [LN2]
Beilliard, Yann [Auteur]
Institut Interdisciplinaire d'Innovation Technologique [Sherbrooke] [3IT]
Laboratoire Nanotechnologies et Nanosystèmes [Sherbrooke] [LN2]

Titre de la revue :

Frontiers in Electronics

Pagination :

825077

Éditeur :

Frontiers Media

Date de publication :

2022-03-25

Mot(s)-clé(s) en anglais :

resistive switching memories
memristor
in-memory computing
hardware non-idealities
artificial neural networks
bayesian neural networks
probabilistic computing

Discipline(s) HAL :

Sciences de l'ingénieur [physics]

Résumé en anglais : [en]

Novel computing architectures based on resistive switching memories (also known as memristors or RRAMs) have been shown to be promising approaches for tackling the energy inefficiency of deep learning and spiking neural ...
Lire la suite >Novel computing architectures based on resistive switching memories (also known as memristors or RRAMs) have been shown to be promising approaches for tackling the energy inefficiency of deep learning and spiking neural networks. However, resistive switch technology is immature and suffers from numerous imperfections, which are often considered limitations on implementations of artificial neural networks. Nevertheless, a reasonable amount of variability can be harnessed to implement efficient probabilistic or approximate computing. This approach turns out to improve robustness, decrease overfitting and reduce energy consumption for specific applications, such as Bayesian and spiking neural networks. Thus, certain non-idealities could become opportunities if we adapt machine learning methods to the intrinsic characteristics of resistive switching memories. In this short review, we introduce some key considerations for circuit design and the most common non-idealities. We illustrate the possible benefits of stochasticity and compression with examples of well-established software methods. We then present an overview of recent neural network implementations that exploit the imperfections of resistive switching memory, and discuss the potential and limitations of these approaches.Lire moins >

Langue :

Anglais

Vulgarisation :

Non

Projet ANR :

Unsupervised spiking neural networks with analog memristive devices for edge computing

Collections :

• Institut d'Électronique, de Microélectronique et de Nanotechnologie (IEMN) - UMR 8520

Source :

Harvested from HAL