Abstract
We present and experimentally evaluate the use of transfer learning to address experimental data scarcity when training neural network (NN) models for Mach–Zehnder interferometer mesh-based optical matrix multipliers. Our approach involves pretraining the model using synthetic data generated from a less accurate analytical model and fine-tuning it with experimental data. Our investigation demonstrates that this method yields significant reductions in modeling errors compared to using an analytical model or a standalone NN model when training data is limited. Utilizing regularization techniques and ensemble averaging, we achieve <1 dB root-mean-square error on the 3×3 matrix weights implemented by a photonic chip while using only $25{\% }$ of the available data.
© 2023 Optica Publishing Group
Full Article | PDF ArticleMore Like This
Lareb Zar Khan, João Pedro, Nelson Costa, Andrea Sgambelluri, Antonio Napoli, and Nicola Sambo
J. Opt. Commun. Netw. 16(3) 369-381 (2024)
Zhedong Wang, Min Chen, Chao Qian, Zhixiang Fan, Huaping Wang, and Hongsheng Chen
Opt. Lett. 47(22) 5897-5900 (2022)
Zhiwei Tang, Fei Wang, ZhenFeng Fu, Shanshan Zheng, Ying Jin, and Guohai Situ
Opt. Lett. 48(9) 2285-2288 (2023)