Photonic Neural Network Fabricated on Thin Film Lithium Niobate for High-Fidelity and Power-Efficient Matrix Computation
We present here the first implementation of a photonic neural network based on the thin-film lithium niobate platform. Our device features ultra-high fidelity as 98.5% and exceptional power efficiency as 33.5 fJ per operation.
Authors
- Yong Zheng (ORCID: https://orcid.org/0000-0002-9124-6035)
- Min Wang (ORCID: https://orcid.org/0000-0001-5321-3323)
- Yuan Ren (ORCID: https://orcid.org/0000-0002-6373-2685)
- Ya Cheng (ORCID: https://orcid.org/0000-0001-8269-5907)
- Jian Liu (ORCID: https://orcid.org/0000-0003-0756-2260)
- Rongbo Wu (ORCID: https://orcid.org/0000-0001-8012-0456)
- Rui Bao (ORCID: https://orcid.org/0009-0007-7876-8115)
- Yu Ma (ORCID: https://orcid.org/0000-0003-4360-2098)
Institutions
- Chinese Academy of Sciences (CN)
- Shanghai Institute of Optics and Fine Mechanics (CN)
- State Key Laboratory of High Field Laser Physics
- East China Normal University (CN)
Publication Details
- Published
- 2024-08-04
- DOI
- https://doi.org/10.1109/cleo-pr60912.2024.10676772
- Citations
- 1
- Primary Topic
- Photonic and Optical Devices
- Type
- article
- Field-Weighted Citation Impact
- 0.21