Scalable Optical Learning Operator

TL;DR

This paper introduces an optical computing framework using multimode fibers that enables high-speed, energy-efficient machine learning tasks such as image classification and speech recognition, matching digital accuracy.

Contribution

It presents a novel optical learning operator leveraging multimode fiber effects, overcoming energy scaling issues while maintaining high speed and accuracy.

Findings

Successfully classified COVID-19 X-ray images

Achieved speech recognition comparable to digital methods

Predicted age from face images with high accuracy

Abstract

Today's heavy machine learning tasks are fueled by large datasets. Computing is performed with power hungry processors whose performance is ultimately limited by the data transfer to and from memory. Optics is one of the powerful means of communicating and processing information and there is intense current interest in optical information processing for realizing high-speed computations. Here we present and experimentally demonstrate an optical computing framework based on spatiotemporal effects in multimode fibers for a range of learning tasks from classifying COVID-19 X-ray lung images and speech recognition to predicting age from face images. The presented framework overcomes the energy scaling problem of existing systems without compromising speed. We leveraged simultaneous, linear, and nonlinear interaction of spatial modes as a computation engine. We numerically and experimentally showed the ability of the method to execute several different tasks with accuracy comparable to a digital implementation.