Diffusion-Based Electromagnetic Inverse Design of Scattering Structured Media

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Tsukerman, Mikhail, Grotov, Konstantin, Ginzburg, Pavel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908636104622080
author Tsukerman, Mikhail
Grotov, Konstantin
Ginzburg, Pavel
author_facet Tsukerman, Mikhail
Grotov, Konstantin
Ginzburg, Pavel
contents We present a conditional diffusion model for electromagnetic inverse design that generates structured media geometries directly from target differential scattering cross-section profiles, bypassing expensive iterative optimization. Our 1D U-Net architecture with Feature-wise Linear Modulation learns to map desired angular scattering patterns to 2x2 dielectric sphere structure, naturally handling the non-uniqueness of inverse problems by sampling diverse valid designs. Trained on 11,000 simulated metasurfaces, the model achieves median MPE below 19% on unseen targets (best: 1.39%), outperforming CMA-ES evolutionary optimization while reducing design time from hours to seconds. These results demonstrate that employing diffusion models is promising for advancing electromagnetic inverse design research, potentially enabling rapid exploration of complex metasurface architectures and accelerating the development of next-generation photonic and wireless communication systems. The code is publicly available at https://github.com/mikzuker/inverse_design_metasurface_generation.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05357
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Diffusion-Based Electromagnetic Inverse Design of Scattering Structured Media
Tsukerman, Mikhail
Grotov, Konstantin
Ginzburg, Pavel
Machine Learning
Applied Physics
Computational Physics
We present a conditional diffusion model for electromagnetic inverse design that generates structured media geometries directly from target differential scattering cross-section profiles, bypassing expensive iterative optimization. Our 1D U-Net architecture with Feature-wise Linear Modulation learns to map desired angular scattering patterns to 2x2 dielectric sphere structure, naturally handling the non-uniqueness of inverse problems by sampling diverse valid designs. Trained on 11,000 simulated metasurfaces, the model achieves median MPE below 19% on unseen targets (best: 1.39%), outperforming CMA-ES evolutionary optimization while reducing design time from hours to seconds. These results demonstrate that employing diffusion models is promising for advancing electromagnetic inverse design research, potentially enabling rapid exploration of complex metasurface architectures and accelerating the development of next-generation photonic and wireless communication systems. The code is publicly available at https://github.com/mikzuker/inverse_design_metasurface_generation.
title Diffusion-Based Electromagnetic Inverse Design of Scattering Structured Media
topic Machine Learning
Applied Physics
Computational Physics
url https://arxiv.org/abs/2511.05357