Dimensionality Reduction in the Design Domain of Photonic Crystal Waveguides for Deep Neural Networks by Implementing Transfer Learning
In this study, transfer learning was implemented to reduce the dimensionality in the nanophotonic design domain. As a result, we could train the network for the 2D photonic crystal devices with the 1D photonic crystal devices dataset.
Authors
- Berkay Neşeli (ORCID: https://orcid.org/0000-0001-6580-653X)
- Hamza Kurt (ORCID: https://orcid.org/0000-0002-0749-4205)
- Junhyeong Kim (ORCID: https://orcid.org/0000-0001-6855-3240)
- Hyo-Hoon Park
Institutions
- Korea Advanced Institute of Science and Technology (KR)
Publication Details
- Published
- 2024-08-04
- DOI
- https://doi.org/10.1109/cleo-pr60912.2024.10676672
- Primary Topic
- Photonic Crystals and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00