Inverse Design Local-Density-of-States via Deep Learning in Quantum Nanophotonics

Abstract

The inverse design method based on deep learning subverts the field of nanophotonics. The nanophotonic system that quantum emitters couple with a nanophotonic structure obtains excellent attention since it can introduce the light-matter interactions to quantum nanophotonics. The light-matter interactions are usually described by a scalar quantity, local-density-of-states. In this manuscript, we apply a fully-connected neural network to approximate the local-density-of-states of the quantum nanophotonic system consisting of a multilayer shell metallic nanoparticle and a quantum emitter for the first time. In addition, we propose the loss function for inverse design to manipulate the resonant frequency, its amplitude, and its linewidth simultaneously. Our work introduces deep learning to the quantum optics domain for advancing quantum device designs; and provides a new platform for practicing deep learning to design nanophotonic structures.