logo
ResearchBunny Logo
Powering AI at the edge: A robust, memristor-based binarized neural network with near-memory computing and miniaturized solar cell

Engineering and Technology

Powering AI at the edge: A robust, memristor-based binarized neural network with near-memory computing and miniaturized solar cell

F. Jebali, A. Majumdar, et al.

Discover the groundbreaking work by Fadi Jebali and colleagues on memristor-based neural networks that can power AI autonomously using energy harvesters. Their innovative approach utilizes a miniature solar cell to enable digital near-memory computing that adapts to varying lighting conditions. This research promises energy-efficient solutions for intelligent sensors in various applications.

00:00
00:00
Playback language: English
Abstract
Memristor-based neural networks offer energy-efficient AI, potentially self-powered with energy harvesters. However, most memristor networks rely on analog in-memory computing, needing stable power incompatible with unstable energy harvesters. This work details a robust binarized neural network (32,768 memristors) powered by a miniature wide-bandgap solar cell. Using digital near-memory computing with complementary memristors and logic-in-sense-amplifier, the circuit avoids compensation/calibration, functioning under diverse conditions. High illumination yields performance comparable to a lab power supply; low illumination maintains functionality with slightly reduced accuracy, transitioning to approximate computing. Image classification simulations show that misclassifications under low illumination mainly involve difficult-to-classify images. This approach enables self-powered AI and intelligent sensors for various applications.
Publisher
Nature Communications
Published On
Jan 25, 2024
Authors
Fadi Jebali, Atreya Majumdar, Clément Turck, Kamel-Eddine Harabi, Mathieu-Coumba Faye, Eloi Muhr, Jean-Pierre Walder, Oleksandr Bilousov, Amadéo Michaud, Elisa Vianello, Tifenn Hirtzlin, François Andrieu, Marc Bocquet, Stéphane Collin, Damien Querlioz, Jean-Michel Portal
Tags
memristor
neural networks
energy harvesting
solar cell
digital computing
self-powered AI
low illumination
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny