Annealing-inspired training of an optical neural network with ternary weights

TL;DR

This paper introduces a ternary weight optical neural network using semiconductor lasers, along with a novel in-situ optimization algorithm, demonstrating high stability and improved convergence suited for resource-constrained hardware.

Contribution

It presents a new ternary weight architecture for optical neural networks and a compatible in-situ optimization algorithm, enhancing efficiency and stability in photonic hardware implementations.

Findings

Achieved ternary weights with Boolean hardware.

The optimization algorithm improves convergence speed and performance.

The ONN maintains over 99% inference stability for more than 10 hours.

Abstract

Artificial neural networks (ANNs) represent a fundamentally connectionnist and distributed approach to computing, and as such they differ from classical computers that utilize the von Neumann architecture. This has revived research interest in new unconventional hardware to enable more efficient implementations of ANNs rather than emulating them on traditional machines. In order to fully leverage the capabilities of this new generation of ANNs, optimization algorithms that take into account hardware limitations and imperfections are necessary. Photonics represents a particularly promising platform, offering scalability, high speed, energy efficiency, and the capability for parallel information processing. Yet, fully fledged implementations of autonomous optical neural networks (ONNs) with in-situ learning remain scarce. In this work, we propose a ternary weight architecture high-dimensional semiconductor laser-based ONN. We introduce a simple method for achieving ternary weights with Boolean hardware, significantly increasing the ONN's information processing capabilities. Furthermore, we design a novel in-situ optimization algorithm that is compatible with, both, Boolean and ternary weights, and provide a detailed hyperparameter study of said algorithm for two different tasks. Our novel algorithm results in benefits, both in terms of convergence speed and performance. Finally, we experimentally characterize the long-term inference stability of our ONN and find that it is extremely stable with a consistency above 99\% over a period of more than 10 hours, addressing one of the main concerns in the field. Our work is of particular relevance in the context of in-situ learning under restricted hardware resources, especially since minimizing the power consumption of auxiliary hardware is crucial to preserving efficiency gains achieved by non-von Neumann ANN implementations.