Predicting local plasmon resonances and geometries using autoencoder networks in complex nanoparticle assemblies

Developing nanostructures with desired optical properties is critical to the field of nanophotonics. To pave the way towards stochastic design of nanoplasmonic structures, we establish the correlative relationship between local nanoparticle geometries and their plasmonic responses. By using an encoder-decoder neural networks within the framework of PyTorch, we demonstrate the predictions of spectra from geometries in the im2spec network, as well as the predictions of geometry from spectral inputs in the spec2im network Equipped we an electron energy (EEL) acquired high angle annular dark field image of a self-assembled monolayer of fluorine and doped indium oxide nanocrystal the HAADF and associating an EEL we image-spectrum pairs are to the network the correlative relationship between each. Each the images (spectra) to a small number of latent variables in a bottleneck fashion, then decodes the latent variables into spectra (images). reduced descriptions contained in the so-called latent can yield a surprising insight the generative mechanisms of complicated