Designing nanophotonic structures using conditional deep convolutional generative adversarial networks

TLDR

Deep‑learning approaches have been introduced in nanophotonics to reduce time‑consuming iterative simulations, a major challenge. The study reports the first use of conditional deep convolutional generative adversarial networks to design unconstrained nanophotonic antennae. The network generates design images from input reflection spectra, enabling suggestions of novel structures beyond parameterized forms. Simulations of the generated designs match the target reflection spectra, demonstrating a fast, convenient method for designing complex nanophotonic structures with desired optical properties.

Abstract

Abstract Data-driven design approaches based on deep learning have been introduced in nanophotonics to reduce time-consuming iterative simulations, which have been a major challenge. Here, we report the first use of conditional deep convolutional generative adversarial networks to design nanophotonic antennae that are not constrained to predefined shapes. For given input reflection spectra, the network generates desirable designs in the form of images; this allows suggestions of new structures that cannot be represented by structural parameters. Simulation results obtained from the generated designs agree well with the input reflection spectrum. This method opens new avenues toward the development of nanophotonics by providing a fast and convenient approach to the design of complex nanophotonic structures that have desired optical properties.