Learning qBIC Resonances across Metasurface Families in Dielectric Fourier Space
The study developed a method to learn qBIC resonances across seven different dielectric metasurface families by mapping them onto a shared reciprocal-lattice grid in Fourier space. This allows for cross-geometry comparison and accurate prediction of resonance shifts. The resulting model significantly reduced error in predicting resonance positions and depths, establishing a unified design space for high-Q resonant photonic structures.