Hybrid training of optical neural networks

TL;DR

This paper introduces a hybrid training method for optical neural networks that trains weights using optical neuron activations, improving robustness against physical imperfections and bridging the reality gap.

Contribution

It presents a novel hybrid training scheme that enables in-situ training of optical neural networks using optical forward propagation, applicable to various architectures.

Findings

Hybrid training is robust against static noise.

The method is applicable to different optical network types.

It bridges the gap between simulation and physical implementation.

Abstract

Optical neural networks are emerging as a promising type of machine learning hardware capable of energy-efficient, parallel computation. Today's optical neural networks are mainly developed to perform optical inference after in silico training on digital simulators. However, various physical imperfections that cannot be accurately modelled may lead to the notorious reality gap between the digital simulator and the physical system. To address this challenge, we demonstrate hybrid training of optical neural networks where the weight matrix is trained with neuron activation functions computed optically via forward propagation through the network. We examine the efficacy of hybrid training with three different networks: an optical linear classifier, a hybrid opto-electronic network, and a complex-valued optical network. We perform a comparative study to in silico training, and our results show that hybrid training is robust against different kinds of static noise. Our platform-agnostic hybrid training scheme can be applied to a wide variety of optical neural networks, and this work paves the way towards advanced all-optical training in machine intelligence.