Reinforcement Learning in a large scale photonic Recurrent Neural Network
This work addresses the problem of realizing efficient, parallel learning hardware for photonic machine learning substrates, which could be disruptive for future technology, though it appears incremental as it builds on existing photonic neural network concepts.
The researchers tackled the challenge of implementing learning in large-scale photonic neural networks by demonstrating a network of up to 2500 diffractively coupled photonic nodes using reinforcement learning, achieving efficient convergence and very good performance.
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.