Reinforcement learning in a large-scale photonic recurrent neural network

J. Bueno, S. Maktoobi, L. Froehly, I. Fischer, M. Jacquot, L. Larger, D. Brunner

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Photonic neural network implementation has 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 has been lacking so far. We demonstrate a network of up to 2025 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.

LanguageEnglish
Pages756-760
Number of pages5
JournalOptica
Volume5
Issue number6
DOIs
Publication statusPublished - 20 Jun 2018

Fingerprint

Recurrent neural networks
Reinforcement learning
reinforcement
Photonics
learning
photonics
Neural networks
machine learning
Energy efficiency
Learning systems
Mirrors
Hardware
Bandwidth
hardware
mirrors
bandwidth
Substrates
output
energy

Keywords

  • neural networks
  • nonlinear optics
  • optical neural systems

Cite this

Bueno, J., Maktoobi, S., Froehly, L., Fischer, I., Jacquot, M., Larger, L., & Brunner, D. (2018). Reinforcement learning in a large-scale photonic recurrent neural network. Optica, 5(6), 756-760. https://doi.org/10.1364/OPTICA.5.000756
Bueno, J. ; Maktoobi, S. ; Froehly, L. ; Fischer, I. ; Jacquot, M. ; Larger, L. ; Brunner, D. / Reinforcement learning in a large-scale photonic recurrent neural network. In: Optica. 2018 ; Vol. 5, No. 6. pp. 756-760.
@article{26f3ce40cce24096aeb89ddc9865d442,
title = "Reinforcement learning in a large-scale photonic recurrent neural network",
abstract = "Photonic neural network implementation has 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 has been lacking so far. We demonstrate a network of up to 2025 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.",
keywords = "neural networks, nonlinear optics, optical neural systems",
author = "J. Bueno and S. Maktoobi and L. Froehly and I. Fischer and M. Jacquot and L. Larger and D. Brunner",
year = "2018",
month = "6",
day = "20",
doi = "10.1364/OPTICA.5.000756",
language = "English",
volume = "5",
pages = "756--760",
journal = "Optica Applicata",
issn = "0078-5466",
publisher = "Optical Society of America",
number = "6",

}

Bueno, J, Maktoobi, S, Froehly, L, Fischer, I, Jacquot, M, Larger, L & Brunner, D 2018, 'Reinforcement learning in a large-scale photonic recurrent neural network' Optica, vol. 5, no. 6, pp. 756-760. https://doi.org/10.1364/OPTICA.5.000756

Reinforcement learning in a large-scale photonic recurrent neural network. / Bueno, J.; Maktoobi, S.; Froehly, L.; Fischer, I.; Jacquot, M.; Larger, L.; Brunner, D.

In: Optica, Vol. 5, No. 6, 20.06.2018, p. 756-760.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Reinforcement learning in a large-scale photonic recurrent neural network

AU - Bueno, J.

AU - Maktoobi, S.

AU - Froehly, L.

AU - Fischer, I.

AU - Jacquot, M.

AU - Larger, L.

AU - Brunner, D.

PY - 2018/6/20

Y1 - 2018/6/20

N2 - Photonic neural network implementation has 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 has been lacking so far. We demonstrate a network of up to 2025 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.

AB - Photonic neural network implementation has 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 has been lacking so far. We demonstrate a network of up to 2025 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.

KW - neural networks

KW - nonlinear optics

KW - optical neural systems

UR - http://www.scopus.com/inward/record.url?scp=85048748549&partnerID=8YFLogxK

U2 - 10.1364/OPTICA.5.000756

DO - 10.1364/OPTICA.5.000756

M3 - Article

VL - 5

SP - 756

EP - 760

JO - Optica Applicata

T2 - Optica Applicata

JF - Optica Applicata

SN - 0078-5466

IS - 6

ER -

Bueno J, Maktoobi S, Froehly L, Fischer I, Jacquot M, Larger L et al. Reinforcement learning in a large-scale photonic recurrent neural network. Optica. 2018 Jun 20;5(6):756-760. https://doi.org/10.1364/OPTICA.5.000756