Deep Learning-Enabled Electromagnetic Near-Field Prediction and Inverse Design of Metasurfaces

We introduced deep neural network designs to predict the complete phase map information for a high-DOF inter-coupled metasurface geometry (forward problem) and to find the metasurface geometry from a given EM phase profile (inverse design).

This website provides the source code to train and test the networks. Additionally, it provides access to our training sets and the final network files after they are trained.

NOTE: The following source codes are provided for research purposes only. The authors have no responsibility for any consequences of use of these source codes. If you use any part of these codes, please cite the following paper.

• T. B. Kanmaz, E. Ozturk, H. V. Demir, and C. Gunduz-Demir, “Deep learning-enabled electromagnetic near-field prediction and inverse design of metasurfaces,” Optica, 2023 (accepted).

Please contact Tevfik Bulent Kanmaz for further questions.