Deep learning enables large-scale inverse design of free-form metasurfaces

arXiv:2610.05021 · physics.optics, cond-mat.mes-hall, physics.comp-ph · Submitted 2026-10-04 · Read on arXiv

physics.optics, cond-mat.mes-hall, physics.comp-ph

Submitted: 2026-10-04

Updated: 2026-10-04

Terminology

• Inverse Design — 1× in this paper · explained in 2 episodes

Sources

• Physics-informed Neural Operators for Predicting 3D Electromagnetic Fields Transformed by Metasurfaces
• Design and Optimization of Metasurfaces for Silicon Photonics: PhD Thesis
• Symmetry-Informed Deep Learning for Electromagnetic Scattering

Related papers

• Quantitative Benchmarking of Spectroscopic Homogeneous and Inhomogeneous Linewidth Separation
• Reciprocal asymmetric transmission in self-shadowed metallized gratings
• Method for SOFI-based spatial super-resolution in nanosensing with blinking emitters
• Confocal imaging from biphoton correlations
• Flexible Inverse Design of Optical Multilayer Coatings Using Discrete-Continuous Flow Matching
• Quantum-Limited Optical Vector Analysis