Figure 2
From: Deep learning based analysis of microstructured materials for thermal radiation control
(a) neural network predictions of the optical properties compared to the properties obtained from FDTD simulations plotted for the test dataset. (b) By material average mean absolute error (MAE) for reflectivity (orange) and transmissivity (blue) separately for the test dataset. No error exceeds 0.01, the overall test dataset has an MAE of 0.0034. (c) The extinction coefficient (k) vs the refractive index (n) for all of the materials included in the test dataset, highlighting the differences between the materials used in the training of the network.
