Multiscale Physics-Informed Neural Networks for the Inverse Design of Hyperuniform Optical Materials
arXiv:2405.07878 · doi:10.1002/adom.202403304
Abstract
In this article, we employ multiscale physics-informed neural networks (MscalePINNs) for the inverse design of finite-size photonic materials with stealthy hyperuniform (SHU) disordered geometries. Specifically, we show that MscalePINNs can capture the fast spatial variations of complex fields scattered by arrays of dielectric nanocylinders arranged according to isotropic SHU point patterns, thus enabling a systematic methodology to inversely retrieve their effective dielectric profiles. Our approach extends the recently developed high-frequency homogenization theory of hyperuniform media and retrieves more general permittivity profiles for applications-relevant finite-size SHU systems, unveiling unique features related to their isotropic nature. In particular, we numerically corroborate the existence of a transparency region beyond the long-wavelength approximation, enabling effective and isotropic homogenization even without disorder-averaging, in contrast to the case of uncorrelated Poisson random patterns. The flexible multiscale network approach introduced here enables the efficient inverse design of more general effective media and finite-size optical metamaterials with isotropic electromagnetic responses beyond the limitations of traditional homogenization theories.
References in corpus (27)
- A high-bias, low-variance introduction to Machine Learning for physicists
- Local Density Fluctuations, Hyperuniformity, and Order Metrics
- On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networks
- Physics-informed neural networks for inverse problems in nano-optics and metamaterials
- Deep neural networks for the evaluation and design of photonic devices
- Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
- Hyperuniform States of Matter
- Designer disordered materials with large complete photonic band gaps
- Photon Management in Two-Dimensional Disordered Media
- Anomalous Localization in Low-Dimensional Systems with Correlated Disorder
- Deep learning meets nanophotonics: A generalized accurate predictor for near fields and far fields of arbitrary 3D nanostructures
- Inverse design in photonics by topology optimization: tutorial
- Isotropic Band Gaps and Freeform Waveguides Observed in Hyperuniform Disordered Photonic Solids
- Machine-Learning-Assisted Metasurface Design for High-Efficiency Thermal Emitter Optimization
- Efficient and accurate inversion of multiple scattering with deep learning
- Designing Disordered Hyperuniform Two-Phase Materials with Novel Physical Properties
- Hyperuniform disordered phononic structures
- Random Scalar Fields and Hyperuniformity
- Physics-informed neural networks for imaging and parameter retrieval of photonic nanostructures from near-field data
- Multifunctional Composites for Elastic and Electromagnetic Wave Propagation
- Experimental Tuning of Transport Regimes in Hyperuniform Disordered Photonic Materials
- Sub-diffusive wave transport and weak localization transition in three-dimensional stealthy hyperuniform disordered systems
- Designing Collective Non-local Responses of Metasurfaces
- Theoretical Prediction of the Effective Dynamic Dielectric Constant of Disordered Hyperuniform Anisotropic Composites Beyond the Long-Wavelength Regime
- Auxiliary Physics-Informed Neural Networks for Forward, Inverse, and Coupled Radiative Transfer Problems
- Designing Multi-functional Metamaterials
- Inverse design of functional photonic patches by adjoint optimization coupled to the generalized Mie theory