Abstract
We propose a new neural network approach to simulating light propagation using direct intermediate prediction, which requires fewer independent training simulations, admits an efficiently-calculated physics-informed loss function, and produces more accurate predictions than existing methods.
© 2023 The Author(s)
PDF ArticleMore Like This
Sonia Boscolo, John M. Dudley, and Christophe Finot
SM4F.7 CLEO: Science and Innovations (CLEO:S&I) 2023
M. Hossein Eybposh, Changjia Cai, Vincent R. Curtis, and Nicolas C. Pegard
1272325 Education and Training in Optics and Photonics (ETOP) 2023
Y. C. Zhan, H. Zhang, H. Cai, D. P. Poenar, L. C. Kwek, and A. Q. Liu
JTh2A.22 CLEO: Applications and Technology (CLEO:A&T) 2023