Laser interferometry as a robust neuromorphic platform for machine learning

TL;DR

This paper introduces a linear optical neural network platform using laser interferometry that enables efficient in situ training and inference, demonstrating robustness to photon losses and simplifying experimental implementation.

Contribution

It presents a novel optical neural network method leveraging linear optics and phase encoding for in situ training and inference, advancing experimental feasibility.

Findings

The proposed optical neural network is resilient to photon losses.

In situ training can be performed using established gradient techniques.

The method simplifies experimental implementation compared to previous approaches.

Abstract

We present a method for implementing an optical neural network using only linear optical resources, namely field displacement and interferometry applied to coherent states of light. The nonlinearity required for learning in a neural network is realized via an encoding of the input into phase shifts allowing for far more straightforward experimental implementation compared to previous proposals for, and demonstrations of, $\textit{in situ}$ inference. Beyond $\textit{in situ}$ inference, the method enables $\textit{in situ}$ training by utilizing established techniques like parameter shift methods or physical backpropagation to extract gradients directly from measurements of the linear optical circuit. We also investigate the effect of photon losses and find the model to be very resilient to these.