Photonic Advantage of Optical Encoders

  • Luocheng Huang
  • , Saswata Mukherjee
  • , Quentin Tanguy
  • , Johannes Froch
  • , Arka Majumdar

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

1 Scopus citations

Abstract

This research presents an optical/digital hybrid ANN which demonstrates a "photonic advantage" in intermediate classification accuracy over pure electronic ANNs with the same power and latency. The optical encoder uses incoherent light which is suitable for operation in ambient light with no additional optical power needed. Further research is needed to explore its efficiency in more complicated datasets, training algorithms, and nonlinear activation in the optical domain.

Original languageEnglish
Title of host publication2023 Conference on Lasers and Electro-Optics, CLEO 2023
PublisherInstitute of Electrical and Electronics Engineers
ISBN (Electronic)9781957171258
StatePublished - 1 Jan 2023
Externally publishedYes
Event2023 Conference on Lasers and Electro-Optics, CLEO 2023 - San Jose, United States
Duration: 7 May 202312 May 2023

Publication series

Name2023 Conference on Lasers and Electro-Optics, CLEO 2023

Conference

Conference2023 Conference on Lasers and Electro-Optics, CLEO 2023
Country/TerritoryUnited States
CitySan Jose
Period7/05/2312/05/23

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Electronic, Optical and Magnetic Materials
  • Instrumentation
  • Atomic and Molecular Physics, and Optics

Fingerprint

Dive into the research topics of 'Photonic Advantage of Optical Encoders'. Together they form a unique fingerprint.

Cite this