Low-power scalable multilayer optoelectronic neural networks enabled with incoherent light

Optical approaches have made great strides towards the goal of high-speed, energy-efficient computing necessary for modern deep learning and AI applications. Read-in and read-out of data, however, limit the overall performance of existing approaches. This study introduces a multilayer optoelectronic...

Full description

Saved in:
Bibliographic Details
Published in:Nature communications 2024-12, Vol.15 (1), p.10692-12, Article 10692
Main Authors: Song, Alexander, Murty Kottapalli, Sai Nikhilesh, Goyal, Rahul, Schölkopf, Bernhard, Fischer, Peer
Format: Article
Language:English
Subjects:
639/624
639/705
Artificial neural networks
Computation
Deep learning
Energy efficiency
Humanities and Social Sciences
Linear functions
Machine learning
Matrices (mathematics)
multidisciplinary
Multilayers
Neural networks
Optoelectronics
Real time operation
Science
Science (multidisciplinary)
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Optical approaches have made great strides towards the goal of high-speed, energy-efficient computing necessary for modern deep learning and AI applications. Read-in and read-out of data, however, limit the overall performance of existing approaches. This study introduces a multilayer optoelectronic computing framework that alternates between optical and optoelectronic layers to implement matrix-vector multiplications and rectified linear functions, respectively. Our framework is designed for real-time, parallelized operations, leveraging 2D arrays of LEDs and photodetectors connected via independent analog electronics. We experimentally demonstrate this approach using a system with a three-layer network with two hidden layers and operate it to recognize images from the MNIST database with a recognition accuracy of 92% and classify classes from a nonlinear spiral data with 86% accuracy. By implementing multiple layers of a deep neural network simultaneously, our approach significantly reduces the number of read-ins and read-outs required and paves the way for scalable optical accelerators requiring ultra low energy. Read-in and read-out of data limit the overall performance of optical computing methods. This work introduces a multilayer optoelectronic framework that alternates between optical and optoelectronic layers to implement matrix-vector multiplications and rectified linear functions experimentally
ISSN:2041-1723
2041-1723
DOI:10.1038/s41467-024-55139-4