Communication Dans Un Congrès Année : 2019

Reinforcement Learning in a Large Scale Photonic Network

Résumé

We have recently succeeded in the implementation of a large scale recurrent photonic neural network hosting up to 2025 photonic neurons. All network internal and readout connections are physically implemented with fully parallel technology. Based on a digital micro-mirror array, we can train the Boolean readout weights using a greedy version of greedy learning. We find that the learning excellently converges. Furthermore, it appears to possess a conveniently convex-like cost-function and demonstrates exceptional scalability of the learning effort with system size.I will introduce our photonic neural network in detail and give a general motivation of photonic systems for neural network processors. Finally, I will discuss the obtained findings of the learning procedure in light of their relevance for hardware implemented neural networks.

Fichier non déposé

Dates et versions

hal-02382667 , version 1 (27-11-2019)

Identifiants

  • HAL Id : hal-02382667 , version 1

Citer

Daniel Brunner, Louis Andreoli, Sheler Maktoobi, Maxime Jacquot, Laurent Larger. Reinforcement Learning in a Large Scale Photonic Network. Coherent Network Computing, Mar 2019, Atsugi, Japan. ⟨hal-02382667⟩
67 Consultations
1 Téléchargements

Partager

  • More