Sample-efficient inverse design of freeform nanophotonic devices with physics-informed reinforcement learning.

Finding an optimal device structure in the vast combinatorial design space of freeform nanophotonic design has been an enormous challenge. In this study, we propose physics-informed reinforcement learning (PIRL) that combines the adjoint-based method with reinforcement learning to improve the sample efficiency by an order of magnitude compared to conventional reinforcement learning and overcome the issue of local minima. To illustrate these advantages of PIRL over other conventional optimization algorithms, we design a family of one-dimensional metasurface beam deflectors using PIRL, exceeding most reported records.

Design parameters of free-form color splitters for subwavelength pixelated image sensors.

Metasurface-based color splitters are emerging as next-generation optical components for image sensors, replacing classical color filters and microlens arrays. In this work, we report how the design parameters such as the device dimensions and refractive indices of the dielectrics affect the optical efficiency of the color splitters. Also, we report how the design grid resolution parameters affect the optical efficiency and discover that the fabrication of a color splitter is possible even in legacy fabrication facilities with low structure resolutions.

Fast and rigorous optical simulation of periodically corrugated light-emitting diodes based on a diffraction matrix method.

Increasing the light extraction efficiency has been widely studied for highly efficient organic light-emitting diodes (OLEDs). Among many light-extraction approaches proposed so far, adding a corrugation layer has been considered a promising solution for its simplicity and high effectiveness. While the working principle of periodically corrugated OLEDs can be qualitatively explained by the diffraction theory, dipolar emission inside the OLED structure makes its quantitative analysis challenging, making one rely on finite-element electromagnetic simulations that could require huge computing resources.