TR2018-180
Deep Neural Network Inverse Design of Integrated Photonic Power Splitters
-
- , "Deep Neural Network Inverse Design of Integrated Photonic Power Splitters", Nature Scientific Reports, DOI: 10.1038/s41598-018-37952-2, Vol. 9, pp. 1368, December 2018.BibTeX TR2018-180 PDF
- @article{TaherSima2018dec,
- author = {TaherSima, Mohammad and Kojima, Keisuke and Koike-Akino, Toshiaki and Jha, Devesh and Wang, Bingnan and Lin, Chungwei and Parsons, Kieran},
- title = {Deep Neural Network Inverse Design of Integrated Photonic Power Splitters},
- journal = {Nature Scientific Reports},
- year = 2018,
- volume = 9,
- pages = 1368,
- month = dec,
- doi = {10.1038/s41598-018-37952-2},
- issn = {2045-2322},
- url = {https://www.merl.com/publications/TR2018-180}
- }
- , "Deep Neural Network Inverse Design of Integrated Photonic Power Splitters", Nature Scientific Reports, DOI: 10.1038/s41598-018-37952-2, Vol. 9, pp. 1368, December 2018.
-
MERL Contacts:
-
Research Areas:
Artificial Intelligence, Communications, Electronic and Photonic Devices
Abstract:
Predicting physical response of an artificially structured material is of particular interest for scientific and engineering applications. Here we use deep learning to predict optical response of artificially engineered nanophotonic devices. In addition to predicting forward approximation of transmission response for any given topology, this approach allows us to inversely approximate designs for a targeted optical response. Our Deep Neural Network (DNN) could design compact (2.6x2.6 um 2) silicon-on-insulator (SOI)-based 1x2 power splitters with various target splitting ratios in a fraction of a second. This model is trained to minimize the reflection (to smaller than ~ -20 dB) while achieving maximum transmission efficiency above 90% and target splitting specifications. This approach paves the way for rapid design of integrated photonic components relying on complex nanostructures.
Related News & Events
-
NEWS Deep Learning-Based Photonic Circuit Design in Scientific Reports Date: February 4, 2019
Where: Scientific Reports, open-access journal from Nature Research
MERL Contacts: Devesh K. Jha; Toshiaki Koike-Akino; Chungwei Lin; Kieran Parsons; Bingnan Wang
Research Areas: Artificial Intelligence, Electronic and Photonic Devices, Machine LearningBrief- MERL researchers developed a novel design method enhanced by modern deep learning techniques for optimizing photonic integrated circuits (PIC). The developed technique employs residual deep neural networks (DNNs) to understand physics underlaying complicated lightwave propagations through nano-structured photonic devices. It was demonstrated that the trained DNN achieves excellent prediction to design power splitting nanostructures having various target power ratios. The work was published in Scientific Reports, which is an online open access journal from Nature Research, having high-impact articles in the research community.