Deep learning-based design of broadband GHz complex and random metasurfaces

  title={Deep learning-based design of broadband GHz complex and random metasurfaces},
  author={Tianning Zhang and Chun Yun Kee and Yee Sin Ang and Lay Kee Ang},
  journal={APL Photonics},
We are interested to explore the limit in using deep learning (DL) to study the electromagnetic response for complex and random metasurfaces, without any specific applications in mind. For simplicity, we focus on a simple pure reflection problem of a broadband electromagnetic (EM) plane wave incident normally on such complex metasurfaces in the frequency regime of 2 to 12 GHz. In doing so, we create a deep learning (DL) based framework called metasurface design deep convolutional neural network… 


Deep learning modeling approach for metasurfaces with high degrees of freedom.
A deep learning-based metasurface/meta-atom modeling approach is introduced to significantly reduce the characterization time while maintaining accuracy, and features the capability of predicting a meta-atom's wide spectrum response in the timescale of milliseconds.
Deep learning for accelerated all-dielectric metasurface design.
A novel method to solve the inverse modeling problem, termed fast forward dictionary search (FFDS), is developed, which offers tremendous controls to the designer and only requires an accurate forward neural network model.
Phase-to-pattern inverse design paradigm for fast realization of functional metasurfaces via transfer learning
This paper proposes a fast accurate inverse method of designing functional metasurfaces based on transfer learning, which can generate meetasurface patterns monolithically from input phase profiles for specific functions.
Deep Learning: A Rapid and Efficient Route to Automatic Metasurface Design
A triple‐band absorber is designed using the REACTIVE method, where a deep learning model computes the metasurface structure automatically through inputting the desired absorption rate, which convincingly demonstrates the superiority of this design method.
Neural-adjoint method for the inverse design of all-dielectric metasurfaces.
This work proposes and demonstrates a method capable of finding accurate solutions to ill-posed inverse problems, where the conditions of existence and uniqueness are violated, and shows how the neural-adjoint method can intelligently grow the design search space to include designs that increasingly and accurately approximate the desired scattering response.
Plasmonic nanostructure design and characterization via Deep Learning
Rising to the challenge, Haim Suchowski and colleagues from Tel Aviv University in Israel have developed an innovative technique that uses Deep Neural Networks to model the complex relationships between light-matter interactions, allowing them to characterise nanostructures based on their far-field optical responses.
Adaptive Genetic Algorithm for Optical Metasurfaces Design
This paper presents digitized-binary elements, as alternative high-dimensional building blocks, to accommodate the needs of complex-tailorable-multifunctional applications and opens a new gateway of predicting the solution to a problem in the fastest possible way based on statistical analysis of the datasets rather than researching the whole solution space.
Programmable Reflection–Transmission Shared‐Aperture Metasurface for Real‐Time Control of Electromagnetic Waves in Full Space
A “Janus” digital coding metasurface with the capabilities to program various electromagnetic functions in the reflected and transmitted waves simultaneously is presented, validating the full‐space modulations enabled by the programmable metAsurface.
Global optimization of dielectric metasurfaces using a physics-driven neural network
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.
Geometric phase coded metasurface: from polarization dependent directive electromagnetic wave scattering to diffusion-like scattering
The proposed digital metasurfaces provide simple designs and reveal new opportunities for controlling electromagnetic wave scattering with or without polarization dependence.