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

TL;DR

This paper presents a multilayer optoelectronic neural network framework using incoherent light, achieving high-speed, low-power image recognition with reduced data read-in/out, suitable for scalable optical AI accelerators.

Contribution

It introduces a novel multilayer optoelectronic computing system that combines optical and electronic layers for efficient, real-time neural network operations.

Findings

Achieved 92% accuracy on MNIST image recognition

Classified nonlinear spiral data with 86% accuracy

Reduced read-in/read-out operations for scalability

Abstract

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.