Electro-optical Neural Networks based on Time-stretch Method

TL;DR

This paper introduces a novel electro-optical neural network architecture utilizing the time-stretch method, enabling efficient large-scale matrix operations and multi-layer processing with fiber optics, demonstrated on digit recognition with high accuracy.

Contribution

It proposes a new electro-optical neural network design based on time-stretch technology, allowing scalable and efficient optical neural computations.

Findings

Achieved 88% accuracy on handwriting digit recognition

Demonstrated feasibility of multi-layer optical neural networks

Validated performance through numerical simulations

Abstract

In this paper, a novel architecture of electro-optical neural networks based on the time-stretch method is proposed and numerically simulated. By stretching time-domain ultrashort pulses, multiplications of large scale weight matrices and vectors can be implemented on light and multiple-layer of feedforward neural network operations can be easily implemented with fiber loops. Via simulation, the performance of a three-layer electro-optical neural network is tested by the handwriting digit recognition task and the accuracy reaches 88% under considerable noise.