Litcius/Paper detail

General Inverse Design of Layered Thin-Film Materials with Convolutional Neural Networks

Andrew Lininger, Michael Hinczewski, Giuseppe Strangi

2021ACS Photonics63 citationsDOI

Abstract

The design of metamaterials which support unique optical responses is the basis for most thin-film nanophotonic applications. In practice, this inverse design (ID) problem can be difficult to solve systematically due to the large design parameter space associated with general multilayered systems. We apply convolutional neural networks, a subset of deep machine learning, as a tool to solve this ID problem for metamaterials composed of stacks of thin films. We demonstrate the remarkable ability of neural networks to probe the large global design space (up to 1012 possible parameter combinations) and resolve all relationships between the metamaterial structure and corresponding ellipsometric and reflectance/transmittance spectra. The applicability of the approach is further expanded to include the ID of synthetic engineered spectra in general design scenarios. Furthermore, this approach is compared with traditional optimization methods. We find an increase in the relative optimization efficiency of the networks with the increase in the total layer number, revealing the advantage of the machine learning approach in many-layered systems where traditional methods become impractical.

Topics & Concepts

MetamaterialTransmittanceComputer scienceConvolutional neural networkNanophotonicsInverseArtificial neural networkInverse problemParameter spaceOptimization problemBasis (linear algebra)Deep learningMaterials scienceArtificial intelligenceAlgorithmNanotechnologyOptoelectronicsMathematicsMathematical analysisStatisticsGeometryPhotonic Crystals and ApplicationsMetamaterials and Metasurfaces ApplicationsOptical Coatings and Gratings