Deep Learning Meets Nanophotonics: A Generalized Accurate Predictor for Near Fields and Far Fields of Arbitrary 3D Nanostructures

TLDR

Deep neural networks are powerful tools with many potential applications in nanophotonics. The study demonstrates a deep neural network as a fast, general‑purpose predictor of near‑ and far‑field responses of plasmonic and dielectric nanostructures. The authors train a deep neural network and evaluate its strengths and limitations through model studies of single particles and their near‑field interactions. The trained network infers internal fields of arbitrary 3D nanostructures orders of magnitude faster than conventional simulations, accurately reproduces secondary physical quantities, and enables fast, universal design and analysis of nanophotonic systems.

Abstract

Deep artificial neural networks are powerful tools with many possible applications in nanophotonics. Here, we demonstrate how a deep neural network can be used as a fast, general purpose predictor of the full near-field and far-field response of plasmonic and dielectric nanostructures. A trained neural network is shown to infer the internal fields of arbitrary three-dimensional nanostructures many orders of magnitude faster compared to conventional numerical simulations. Secondary physical quantities are derived from the deep learning predictions and faithfully reproduce a wide variety of physical effects without requiring specific training. We discuss the strengths and limitations of the neural network approach using a number of model studies of single particles and their near-field interactions. Our approach paves the way for fast, yet universal, methods for design and analysis of nanophotonic systems.