Synthetic-Domain Neural Networks Using Integrated Nonlinear Phononics on Lithium Niobate

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

Abstract

Leveraging strong second-order nonlinearities of lithium niobate, we demonstrate synthetic-domain neural networks with both efficient linear and nonlinear operations. A high accuracy of 98.2% is experimentally achieved for a four-class classification task.

Original languageEnglish
Title of host publication2025 Conference on Lasers and Electro-Optics, CLEO 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781957171500
StatePublished - 2025
Event2025 Conference on Lasers and Electro-Optics, CLEO 2025 - Long Beach, United States
Duration: May 4 2025May 9 2025

Publication series

Name2025 Conference on Lasers and Electro-Optics, CLEO 2025

Conference

Conference2025 Conference on Lasers and Electro-Optics, CLEO 2025
Country/TerritoryUnited States
CityLong Beach
Period05/4/2505/9/25

Funding

Device fabrication and SEM were conducted as part of user projects (CNMS2022-B-01473, CNMS2024-B-02643) at the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility. Research was sponsored by the Air Force Office of Scientific Research (AFOSR) under W911NF-23-1-0235 and FA9550-22-1-0548 and by Commonwealth Cybersecurity Initiative. The views and conclusions contained in this document are those of the authors and do not necessarily reflect the position or the policy of the United States Government. No official endorsement should be inferred.

Fingerprint

Dive into the research topics of 'Synthetic-Domain Neural Networks Using Integrated Nonlinear Phononics on Lithium Niobate'. Together they form a unique fingerprint.

Cite this