Optoelectronic Loop Neuron
Optoelectronic loop neurons are artificial neural network components designed to mimic biological neural processes using light and superconducting electronics, aiming for energy-efficient and high-speed computation. Current research focuses on developing efficient models and algorithms, including those inspired by cortical loops and direct feedback alignment, to train these networks without relying on computationally expensive backpropagation. This approach leverages the inherent parallelism and low power consumption of photonic and superconducting technologies, potentially leading to significant advancements in neuromorphic computing and applications requiring high-speed, low-power processing.
Papers
May 21, 2024
November 10, 2023
October 18, 2022
June 22, 2022
April 20, 2022