TR2022-047
Finding the Right Deep Neural Network Model for Efficient Design of Tunable Nanophotonic Devices
-
- , "Finding the Right Deep Neural Network Model for Efficient Design of Tunable Nanophotonic Devices", Conference on Lasers and Electro-Optics (CLEO), DOI: 10.1364/CLEO_SI.2022.SW5E.6, May 2022.BibTeX TR2022-047 PDF Video Presentation
- @inproceedings{Jung2022may,
- author = {Jung, Minwoo and Kojima, Keisuke and Koike-Akino, Toshiaki and Wang, Ye and Zhu, Dayu and Brand, Matthew},
- title = {{Finding the Right Deep Neural Network Model for Efficient Design of Tunable Nanophotonic Devices}},
- booktitle = {Conference on Lasers and Electro-Optics (CLEO)},
- year = 2022,
- month = may,
- publisher = {Optica},
- doi = {10.1364/CLEO_SI.2022.SW5E.6},
- isbn = {978-1-957171-05-0},
- url = {https://www.merl.com/publications/TR2022-047}
- }
- , "Finding the Right Deep Neural Network Model for Efficient Design of Tunable Nanophotonic Devices", Conference on Lasers and Electro-Optics (CLEO), DOI: 10.1364/CLEO_SI.2022.SW5E.6, May 2022.
-
MERL Contacts:
-
Research Areas:
Artificial Intelligence, Electronic and Photonic Devices, Machine Learning, Optimization, Signal Processing
Abstract:
We develop generative deep neural networks that explore relevant statistical structures to expedite a complex inverse design of nanophotonic on-chip wavelength demultiplexer. Our design, targeting at telecomm-wavelengths, is electrically switchable via liquid crystal tuning.