Reinforcement Learning in a large scale photonic Recurrent Neural Network

TL;DR

This paper demonstrates a large-scale photonic recurrent neural network with 2500 nodes, utilizing reinforcement learning via a digital micro mirror device, achieving efficient convergence and high performance.

Contribution

It introduces a fully parallel, large-scale photonic RNN with passive weights and reinforcement learning, advancing photonic neural network hardware capabilities.

Findings

Network of 2500 photonic nodes demonstrated.

Reinforcement learning implemented with digital micro mirror device.

Achieved efficient convergence and high performance.

Abstract

Photonic Neural Network implementations have been gaining considerable attention as a potentially disruptive future technology. Demonstrating learning in large scale neural networks is essential to establish photonic machine learning substrates as viable information processing systems. Realizing photonic Neural Networks with numerous nonlinear nodes in a fully parallel and efficient learning hardware was lacking so far. We demonstrate a network of up to 2500 diffractively coupled photonic nodes, forming a large scale Recurrent Neural Network. Using a Digital Micro Mirror Device, we realize reinforcement learning. Our scheme is fully parallel, and the passive weights maximize energy efficiency and bandwidth. The computational output efficiently converges and we achieve very good performance.