Skip to main navigation Skip to search Skip to main content

Thin atomic-vapor cells for all-optical neural network nonlinear activation

  • Roy Maman
  • , David Ohana
  • , Noa Mazurski
  • , Jacob Engelberg
  • , Uriel Levy

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We present thin Rb-vapor cells with nonlinear transmission for compact integration with all-optical neural networks. The efficacy of the cells as nonlinear activation layers is demonstrated experimentally in optical neural networks for image classification.

Original languageEnglish
Title of host publicationCLEO
Subtitle of host publicationScience and Innovations - Proceedings CLEO 2025
PublisherOptical Society of America
ISBN (Electronic)9781957171500
DOIs
StatePublished - 2025
Event2025 CLEO: Science and Innovations, CLEO-SI 2025 - Long Beach, United States
Duration: 4 May 20259 May 2025

Publication series

NameCLEO: Science and Innovations - Proceedings CLEO 2025

Conference

Conference2025 CLEO: Science and Innovations, CLEO-SI 2025
Country/TerritoryUnited States
CityLong Beach
Period4/05/259/05/25

Bibliographical note

Publisher Copyright:
© Optica Publishing Group 2025.

Fingerprint

Dive into the research topics of 'Thin atomic-vapor cells for all-optical neural network nonlinear activation'. Together they form a unique fingerprint.

Cite this