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 (2)

• Ultrasensitive THz sensing with high-Q Fano resonances in metasurfaces
• Freeform metagratings based on complex light scattering dynamics for extreme, high efficiency beam steering

Cited by in corpus (10)

• Machine learning-assisted design of material properties
• Inverse Design of Grating Couplers Using the Policy Gradient Method from Reinforcement Learning
• Deep-Learning-Enabled Inverse Engineering of Multi-Wavelength Invisibility-to-Superscattering Switching with Phase-Change Materials
• Heuristic Methods and Performance Bounds for Photonic Design
• A Freeform Dielectric Metasurface Modeling Approach Based on Deep Neural Networks
• Deep Inverse Design of Reconfigurable Metasurfaces for Future Communications
• Multifunctional Meta-Optic Systems: Inversely Designed with Artificial Intelligence
• Neural-adjoint method for the inverse design of all-dielectric metasurfaces
• 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