raman-lab-ucla Goto Github PK

atmosphere_model

deepadjoint_ml-optimization_integration

Here, we combine deep learning (cDCGAN) with adjoint optimization (LumOpt) within a single user-friendly application (DeepAdjoint)). Reference: https://pubs.acs.org/doi/10.1021/acsphotonics.2c00968

explainability_for_photonics

Here, we use Deep SHAP (or SHAP) to explain the behavior of nanophotonic structures learned by a convolutional neural network (CNN). Reference: https://pubs.acs.org/doi/full/10.1021/acsphotonics.0c01067

ml_tutorials

multiclass_metasurface_inversedesign

Here, we use a conditional deep convolutional generative adversarial network (cDCGAN) to inverse design across multiple classes of metasurfaces. Reference: https://onlinelibrary.wiley.com/doi/10.1002/adom.202100548

scotch_tape_radiative_cooler