TR2020-057
Deep Neural Networks for Designing Integrated Photonics
-
- , "Deep Neural Networks for Designing Integrated Photonics", Optical Fiber Communication Conference and Exposition (OFC), DOI: 10.1364/OFC.2020.Th1A.6, March 2020.BibTeX TR2020-057 PDF
- @inproceedings{Kojima2020mar,
- author = {Kojima, Keisuke and TaherSima, Mohammad and Koike-Akino, Toshiaki and Jha, Devesh K. and Tang, Yingheng and Parsons, Kieran and Sang, Fengqiao and Klamkin, Jonathan},
- title = {Deep Neural Networks for Designing Integrated Photonics},
- booktitle = {Optical Fiber Communication Conference and Exposition (OFC)},
- year = 2020,
- month = mar,
- publisher = {OSA},
- doi = {10.1364/OFC.2020.Th1A.6},
- isbn = {978-1-943580-71-2},
- url = {https://www.merl.com/publications/TR2020-057}
- }
- , "Deep Neural Networks for Designing Integrated Photonics", Optical Fiber Communication Conference and Exposition (OFC), DOI: 10.1364/OFC.2020.Th1A.6, March 2020.
-
MERL Contacts:
-
Research Areas:
Applied Physics, Artificial Intelligence, Communications, Electronic and Photonic Devices, Machine Learning
Abstract:
We present two different approaches to apply deep learning to inverse design for nanophotonic devices. First, we use a regression model, with device parameters as inputs and device responses as outputs, or vice versa. Second, we use a novel generative model to create a series of improved designs. We demonstrate them to design nanophotonic power splitters with multiple splitting ratios