A neural operator-based surrogate solver for free-form electromagnetic inverse design

Abstract

Neural operators have emerged as a powerful tool for solving partial differential equations in the context of scientific machine learning. Here, we implement and train a modified Fourier neural operator as a surrogate solver for electromagnetic scattering problems and compare its data efficiency to existing methods. We further demonstrate its application to the gradient-based nanophotonic inverse design of free-form, fully three-dimensional electromagnetic scatterers, an area that has so far eluded the application of deep learning techniques.

References in corpus (21)

• Decoupled Weight Decay Regularization
• DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
• Gaussian Error Linear Units (GELUs)
• Outlook for inverse design in nanophotonics
• Fourier Neural Operator for Parametric Partial Differential Equations
• A physics-informed variational DeepONet for predicting the crack path in brittle materials
• Adjoint method and inverse design for nonlinear nanophotonic devices
• Deep learning meets nanophotonics: A generalized accurate predictor for near fields and far fields of arbitrary 3D nanostructures
• Inverse design of photonic crystals through automatic differentiation
• DeepM&Mnet: Inferring the electroconvection multiphysics fields based on operator approximation by neural networks
• On the use of Artificial Neural Networks in Topology Optimisation
• Intelligent metaphotonics empowered by machine learning
• Simulator-based training of generative models for the inverse design of metasurfaces
• Benchmarking five global optimization approaches for nano-optical shape optimization and parameter reconstruction
• Active learning of deep surrogates for PDEs: Application to metasurface design
• Forward-Mode Differentiation of Maxwell's Equations
• Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets
• Towards 3D-Printed Inverse-Designed Metaoptics
• Fast multi-source nanophotonic simulations using augmented partial factorization
• Efficient simulation of bi-periodic, layered structures based on the T-matrix method
• Trace formulation for photonic inverse design with incoherent sources

Cited by in corpus (5)

• A newcomer's guide to deep learning for inverse design in nano-photonics
• Deep learning for nano-photonic materials -- The solution to everything!?
• Inverse Design of Unitary Transmission Matrices in Silicon Photonic Coupled Waveguide Arrays using a Neural Adjoint Model
• Data-Efficient Electromagnetic Surrogate Solver Through Dissipative Relaxation Transfer Learning
• Accelerated Time-Domain Simulation of Complex Photonic Structures with a Data-Aware Fourier Neural Operator