Reinforcement Learning in a Large Scale Photonic Network
Sheler Maktoobi, Louis Andréoli, Laurent Larger, Maxime Jacquot, Daniel Brunner · 2018
We experimentally create a neural network, implementing 2025 connections in parallel. We numerically validate the scheme for at least 34000 photonic neurons. We demonstrate photonic efficiently converging reinforcement learning, prediction a chaotic signal with.