Conference paper
Modeling of Optical Matrix Multipliers Using Transposed Convolutional Neural Networks
Department of Electrical and Photonics Engineering, Technical University of Denmark1
Huazhong University of Science and Technology2
Nanophotonic Devices, Department of Electrical and Photonics Engineering, Technical University of Denmark3
Centre of Excellence for Silicon Photonics for Optical Communications, Centers, Technical University of Denmark4
Machine Learning in Photonic Systems, Department of Electrical and Photonics Engineering, Technical University of Denmark5
We demonstrate a data-driven model for optical matrix multipliers utilizing Mach-Zehnder interferometer meshes. For a fabricated chip, a transposed convolutional neural network model learns from experimental measurements offline and predicts the weights across 100 frequency channels in the C-band with high precision (RMSE<0.8 dB).
Language: | English |
---|---|
Publisher: | IEEE |
Year: | 2022 |
Pages: | 1-2 |
Proceedings: | 2022 IEEE Photonics Conference |
ISBN: | 1665434880 , 9781665434881 , 1665434872 and 9781665434874 |
ISSN: | 2575274x |
Types: | Conference paper |
DOI: | 10.1109/IPC53466.2022.9975676 |
ORCIDs: | Cem, Ali , Moura, Uiara Celine De , Ding, Yunhong , Zibar, Darko and Ros, Francesco Da |