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)
Xinlan Ge, Licheng Zhu, Zeyu Gao, Ning Wang, Hongwei Ye, Shuai Wang, and Ping Yang
Opt. Lett. 48(17) 4476-4479 (2023)
Ying Huang, Hengsong Yue, Wei Ma, Yiyuan Zhang, Yao Xiao, Yong Tang, He Tang, and Tao Chu
Opt. Lett. 48(12) 3231-3234 (2023)