Global optimization of dielectric metasurfaces using a physics-driven neural network

Abstract

We present a global optimizer, based on a conditional generative neural network, which can output ensembles of highly efficient topology-optimized metasurfaces operating across a range of parameters. A key feature of the network is that it initially generates a distribution of devices that broadly samples the design space, and then shifts and refines this distribution towards favorable design space regions over the course of optimization. Training is performed by calculating the forward and adjoint electromagnetic simulations of outputted devices and using the subsequent efficiency gradients for backpropagation. With metagratings operating across a range of wavelengths and angles as a model system, we show that devices produced from the trained generative network have efficiencies comparable to or better than the best devices produced by adjoint-based topology optimization, while requiring less computational cost. Our reframing of adjoint-based optimization to the training of a generative neural network applies generally to physical systems that can utilize gradients to improve performance.

20 pages, 5 figures

References in corpus (11)

• Outlook for inverse design in nanophotonics
• Training deep neural networks for the inverse design of nanophotonic structures
• Inverse design and demonstration of a compact and broadband on-chip wavelength demultiplexer
• A Generative Model for Inverse Design of Metamaterials
• Ultrasensitive THz sensing with high-Q Fano resonances in metasurfaces
• Freeform Diffractive Metagrating Design Based on Generative Adversarial Networks
• Metasurface Optics for Full-color Computational Imaging
• Adjoint method and inverse design for nonlinear nanophotonic devices
• Topology Optimized Multi-layered Meta-optics
• RETICOLO software for grating analysis
• Freeform metagratings based on complex light scattering dynamics for extreme, high efficiency beam steering

Cited by in corpus (58)

• Deep neural networks for the evaluation and design of photonic devices
• Deep learning in nano-photonics: inverse design and beyond
• Tunable nanophotonics enabled by chalcogenide phase-change materials
• Freeform Diffractive Metagrating Design Based on Generative Adversarial Networks
• Deep learning meets nanophotonics: A generalized accurate predictor for near fields and far fields of arbitrary 3D nanostructures
• Machine-Learning-Assisted Metasurface Design for High-Efficiency Thermal Emitter Optimization
• Data-Driven Design for Metamaterials and Multiscale Systems: A Review
• Nonlocal Flat Optics
• On the use of Artificial Neural Networks in Topology Optimisation
• Simulator-based training of generative models for the inverse design of metasurfaces
• Enhanced Light-Matter Interactions in Dielectric Nanostructures via Machine Learning Approach
• Intelligent Meta-Imagers: From Compressed to Learned Sensing
• Artificial Neural Networks for Photonic Applications: From Algorithms to Implementation
• Global Inverse Design Across Multiple Photonic Structure Classes Using Generative Deep Learning
• Machine learning-assisted design of material properties
• Diffusion Probabilistic Model Based Accurate and High-Degree-of-Freedom Metasurface Inverse Design
• A Physics Based Approach for Neural Networks Enabled Design of All-Dielectric Metasurfaces
• Hyper Resolution Two Photon Direct Laser Writing using ENZ Nano-Cavity
• Hyper Resolute Ultra thin Low Cost All-Dielectric Broadband Achromatic Metalenses
• A newcomer's guide to deep learning for inverse design in nano-photonics
• General Inverse Design of Thin-Film Metamaterials With Convolutional Neural Networks
• Multi-objective and categorical global optimization of photonic structures based on ResNet generative neural networks
• Neural reparameterization improves structural optimization
• Mapping the Design Space of Photonic Topological States via Deep Learning
• A topological encoding method for data-driven photonics inverse design
• Suppressing meta-holographic artifacts by laser coherence tuning
• A Combined Machine-Learning / Optimization-Based Approach for Inverse Design of Nonuniform Bianisotropic Metasurfaces
• Universal metasurfaces for complete linear control of coherent light transmission
• Machine Learning Framework for Quantum Sampling of Highly-Constrained, Continuous Optimization Problems
• Fast Design of Plasmonic Metasurfaces Enabled by Deep Learning
• Inverse Design of Grating Couplers Using the Policy Gradient Method from Reinforcement Learning
• t-METASET: Tailoring Property Bias of Large-Scale Metamaterial Datasets through Active Learning
• Physics-informed reinforcement learning for sample-efficient optimization of freeform nanophotonic devices
• Deep learning-based design of broadband GHz complex and random metasurfaces
• Deep-Learning-Enabled Inverse Engineering of Multi-Wavelength Invisibility-to-Superscattering Switching with Phase-Change Materials
• Heuristic Methods and Performance Bounds for Photonic Design
• Multifunctional Metasurface Design with a Generative Adversarial Network
• Inverse design with flexible design targets via deep learning: Tailoring of electric and magnetic multipole scattering from nano-spheres
• Predicting nonlinear optical scattering with physics-driven neural networks
• A General Machine Learning-based Approach for Inverse Design of One-dimensional Photonic Crystals Toward Targeted Visible Light Reflection Spectrum
• Machine-Learning-Assisted Photonic Device Development: A Multiscale Approach from Theory to Characterization
• A Freeform Dielectric Metasurface Modeling Approach Based on Deep Neural Networks
• Bridging the reality gap in quantum devices with physics-aware machine learning
• Chat to Chip: Large Language Model Based Design of Arbitrarily Shaped Metasurfaces
• Interpretable inverse-designed cavity for on-chip nonlinear and quantum optics
• Deep Inverse Design of Reconfigurable Metasurfaces for Future Communications
• Physics-guided hierarchical neural networks for Maxwell's equations in plasmonic metamaterials
• Generating extreme quantum scattering in graphene with machine learning
• Predicting VCSEL Emission Properties Using Transformer Neural Networks
• Deep-learning design of graphene metasurfaces for quantum control and Dirac electron holography
• Multifunctional Meta-Optic Systems: Inversely Designed with Artificial Intelligence
• Deep learning for the modeling and inverse design of radiative heat transfer
• Designing metasurface optical interfaces for solid-state qubits using many-body adjoint shape optimization
• Neural-adjoint method for the inverse design of all-dielectric metasurfaces
• A universal deep learning strategy for designing high-quality-factor photonic resonances
• Design space reparameterization enforces hard geometric constraints in inverse-designed nanophotonic devices
• Ultra low loss broadband 1 $\times$ 2 optical power splitters with various splitting ratios
• Spectral splitting and concentration of broadband light using neural networks