A Freeform Dielectric Metasurface Modeling Approach Based on Deep Neural Networks

Metasurfaces have shown promising potentials in shaping optical wavefronts while remaining compact compared to bulky geometric optics devices. Design of meta-atoms, the fundamental building blocks of metasurfaces, relies on trial-and-error method to achieve target electromagnetic responses. This process includes the characterization of an enormous amount of different meta-atom designs with different physical and geometric parameters, which normally demands huge computational resources. In this paper, a deep learning-based metasurface/meta-atom modeling approach is introduced to significantly reduce the characterization time while maintaining accuracy. Based on a convolutional neural network (CNN) structure, the proposed deep learning network is able to model meta-atoms with free-form 2D patterns and different lattice sizes, material refractive indexes and thicknesses. Moreover, the presented approach features the capability to predict meta-atoms' wide spectrum responses in the timescale of milliseconds, which makes it attractive for applications such as fast meta-atom/metasurface on-demand designs and optimizations.

[1]  Jonathan A. Fan,et al.  Global optimization of dielectric metasurfaces using a physics-driven neural network , 2019, Nano letters.

[2]  Tian Gu,et al.  Generative Multi-Functional Meta-Atom and Metasurface Design Networks , 2019, ArXiv.

[3]  Jianjun Hao,et al.  Inverse Design of Photonic Crystal Nanobeam Cavity Structure via Deep Neural Network , 2019, 2019 Asia Communications and Photonics Conference (ACP).

[4]  Wen-Feng Hsieh,et al.  Achieving high numerical aperture near-infrared imaging based on an ultrathin cylinder dielectric metalens. , 2019, Applied optics.

[5]  Yongmin Liu,et al.  Deep-Learning-Enabled On-Demand Design of Chiral Metamaterials. , 2018, ACS nano.

[6]  L. Lewin.  The electrical constants of a material loaded with spherical particles , 1947 .

[7]  Igal Brener,et al.  Polarization-Independent Silicon Metadevices for Efficient Optical Wavefront Control. , 2015, Nano letters.

[8]  Douglas H. Werner,et al.  Optimal High Efficiency 3D Plasmonic Metasurface Elements Revealed by Lazy Ants , 2019, ACS Photonics.

[9]  Sajjad Abdollahramezani,et al.  Knowledge Discovery in Nanophotonics Using Geometric Deep Learning , 2019, ArXiv.

[10]  Wenhua Gu,et al.  Deep neural network for plasmonic sensor modeling , 2019, Optical Materials Express.

[11]  Zhaocheng Liu,et al.  A Hybrid Strategy for the Discovery and Design of Photonic Structures , 2019, IEEE Journal on Emerging and Selected Topics in Circuits and Systems.

[12]  Sergey Levine,et al.  Learning hand-eye coordination for robotic grasping with deep learning and large-scale data collection , 2016, Int. J. Robotics Res..

[13]  Zongfu Yu,et al.  A Bidirectional Deep Neural Network for Accurate Silicon Color Design , 2019, Advanced materials.

[14]  Junsuk Rho,et al.  Designing nanophotonic structures using conditional deep convolutional generative adversarial networks , 2019, Nanophotonics.

[15]  Zhaocheng Liu,et al.  Compounding Meta‐Atoms into Metamolecules with Hybrid Artificial Intelligence Techniques , 2019, Advanced materials.

[16]  A. Arbabi,et al.  Dielectric metasurfaces for complete control of phase and polarization with subwavelength spatial resolution and high transmission. , 2014, Nature nanotechnology.

[17]  Hong Tang,et al.  A Novel Modeling Approach for All-Dielectric Metasurfaces Using Deep Neural Networks , 2019, ArXiv.

[18]  Zongfu Yu,et al.  Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures , 2017, 2019 Conference on Lasers and Electro-Optics (CLEO).

[19]  Feng Cheng,et al.  Probabilistic Representation and Inverse Design of Metamaterials Based on a Deep Generative Model with Semi‐Supervised Learning Strategy , 2019, Advanced materials.

[20]  Yi Yang,et al.  Nanophotonic particle simulation and inverse design using artificial neural networks , 2018, Science Advances.

[21]  Brian A. Slovick,et al.  Generalized effective-medium theory for metamaterials , 2014 .

[22]  Jonathan A. Fan,et al.  Free-Form Di ff ractive Metagrating Design Based on Generative Adversarial Networks , 2019 .

[23]  Danqi Chen,et al.  Reasoning With Neural Tensor Networks for Knowledge Base Completion , 2013, NIPS.

[24]  P. Genevet,et al.  Recent advances in planar optics: from plasmonic to dielectric metasurfaces , 2017 .

[25]  Michael Mrejen,et al.  Plasmonic nanostructure design and characterization via Deep Learning , 2018, Light: Science & Applications.

[26]  Jonathan A. Fan,et al.  Simulator-based training of generative neural networks for the inverse design of metasurfaces , 2019, Nanophotonics.

[27]  W. Cai,et al.  A Generative Model for Inverse Design of Metamaterials , 2018, Nano letters.

[28]  L. Deng,et al.  Ultra-thin high-efficiency mid-infrared transmissive Huygens meta-optics , 2017, Nature Communications.

[29]  Peter R. Wiecha,et al.  Deep learning meets nanophotonics: A generalized accurate predictor for near fields and far fields of arbitrary 3D nanostructures. , 2019, Nano letters.