Communications of the ACM
Artificial Synapses for Neuromorphic Computing
Tested against a dataset of handwritten images from the Modified National Standards and Technology database, the interface-type memristors realized a high image recognition accuracy of 94.72%.
Credit: Los Alamos National Laboratory
Los Alamos National Laboratory (LANL) researchers have developed an interface-type memristive device that could be used to create artificial synapses for next-generation neuromorphic computing.
Memristors can replicate the structure and function of synapses, the human brain's "in-memory processing" system; they also save time and energy by co-locating information storage and processing.
Comprised of a simple structure of gold and other semiconducting materials, the interface-type memristor could be scaled down to nanometer size and needs much less processing power than transistor-based neuromorphic chips.
LANL's Aiping Chen said the advantages of the new device include low-energy consumption, high parallelism, and excellent error tolerance, which ”make it very good for advanced computing tasks like learning, recognition, and decision-making."
From Los Alamos National Laboratory
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, DC, USA
No entries found