Deep recurrent neural networks operations based on stimulated Brillouin scattering
Becker S, Cabello JHM, Englund D, Stiller B (2024)
Publication Type: Conference contribution
Publication year: 2024
Publisher: Optical Society of America
Conference Proceedings Title: CLEO: Science and Innovations, CLEO: S and I 2024 in Proceedings CLEO 2024, Part of Conference on Lasers and Electro-Optics
Event location: Charlotte, NC, USA
Abstract
We experimentally demonstrate eight layers of a recurrent operator for photonic neural networks using stimulated Brillouin scattering. The programmable optoacoustic building block captures and processes coherently and frequency-selectively information without relying on artificial reservoirs.
Authors with CRIS profile
Involved external institutions
Max-Planck-Institut für die Physik des Lichts (MPL) / Max Planck Institute for the Science of Light
Germany (DE)
Massachusetts Institute of Technology (MIT)
United States (USA) (US)
Germany (DE)
Massachusetts Institute of Technology (MIT)
United States (USA) (US)
How to cite
APA:
Becker, S., Cabello, J.H.M., Englund, D., & Stiller, B. (2024). Deep recurrent neural networks operations based on stimulated Brillouin scattering. In CLEO: Science and Innovations, CLEO: S and I 2024 in Proceedings CLEO 2024, Part of Conference on Lasers and Electro-Optics. Charlotte, NC, USA: Optical Society of America.
MLA:
Becker, Steven, et al. "Deep recurrent neural networks operations based on stimulated Brillouin scattering." Proceedings of the CLEO: Science and Innovations in CLEO 2024, CLEO: S and I 2024 - Part of Conference on Lasers and Electro-Optics, Charlotte, NC, USA Optical Society of America, 2024.
BibTeX: Download