Artificial intelligence (AI) techniques have been spreading in most scientific areas and have become a heated focus in photonics research in recent years. Forward modeling and inverse design using AI can achieve high efficiency and accuracy for photonics components. With AI-assisted electronic circuit design for photonics components, more advanced photonics applications have emerged. Photonics benefit a great deal from AI, and AI, in turn, benefits from photonics by carrying out AI algorithms, such as complicated deep neural networks using photonics components that use photons rather than electrons. Beyond the photonics domain, other related research areas or topics governed by Maxwell’s equations share remarkable similarities in using the help of AI. The studies in computational electromagnetics, the design of microwave devices, as well as their various applications greatly benefit from AI. This article reviews leveraging AI in photonics modeling, simulation, and inverse design; leveraging photonics computing for implementing AI algorithms; and leveraging AI beyond photonics topics, such as microwaves and quantum-related topics.

Single-mode operation is crucial in many on-chip integrated photonic devices, and thus the identification of single-mode geometries is an inevitable design requirement. In this article, we develop deep learning (DL) models for ultra-quick classifications of optical waveguide geometries into single- and multi-modal geometries. The DL model accurately predicts the boundary in the parameter space for the geometry of the waveguide that splits the space into single- and multi-modal regions. Using silicon nitride channel waveguide, and targeting both visible and telecommunication wavelengths, we illustrate how DL models can be developed with a minimal number of exact numerical simulations to Maxwell’s equations.

This article displays the method of constructing deep learning models for optical mode solving, with a minimal number of exact numerical solutions to Maxwell’s equations. We select a silicon nitride channel waveguide and show how the patterns in the effective refractive indices of the fundamental waveguide modes for both polarizations of light, can be uncovered with only 4–16 learning points for the entire parameter space that can be conveniently accessed using existing photo-lithographical and CMOS fabrication techniques. We also illustrate the effect of various transfer functions and neural network layouts to the overall performance of the deep learning model.

Light emission from the color centers in diamonds can be significantly enhanced by their interaction with optical microcavities. In the conventional chip-based hybrid approach, nanodiamonds are placed directly on the surface of microcavity chips created using fabrication-matured material platforms. However, the achievable enhancement due to the Purcell effect is limited because of the evanescent interaction between the electrical field of the cavity and the nanodiamond. Here, we propose and statistically analyze a diamond in a nanopocket structure as a new route to achieve a high enhancement of light emission from the color center in the nanodiamond, placed in an optical microcavity. We demonstrate that by creating a nanopocket within the photonic crystal L3 cavity and placing the nanodiamond in, a significant and a robust control over the local density of states can be obtained. The antinodes of the electric field relocate to the nanosized air gaps within the nanopocket, between the nanodiamond and the microcavity. This creates an elevated and uniform electric field across the nanodiamond that is less sensitive to perturbations in the shape and orientation of the nanodiamond. Using a silicon nitride photonic crystal L3 cavity and aiming at silicon-vacancy and nitrogen-vacancy color centers in diamond, we performed a statistical analysis of light emission, assuming random positions of color centers and dipole moment orientations. We showed that in cavities with experimentally feasible quality factors, the diamond in the nanopocket structure produces Purcell factor distributions with mean and median that are tenfold larger compared to what can be achieved when the diamond is on the surface of the microcavity.