Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Neural Adjoint Method for Meta-optics: Accelerating Volumetric Inverse Design via Fourier Neural Operators
Chanik Kang, Hyewon Suk, Haejun Chung
Meta-optics promises compact, high-performance imaging and color routing. However, designing high-performance structures is a high-dimensional optimization problem: mapping a desir…
cs.LG2025
Generalizing Deep Surrogate Solvers for Broadband Electromagnetic Field Prediction at Unseen Wavelengths
Joonhyuk Seo, Chanik Kang, Dongjin Seo +1
Recently, electromagnetic surrogate solvers, trained on solutions of Maxwell's equations under specific simulation conditions, enabled fast inference of computationally expensive s…