Deep learning meets nanophotonics: A generalized Accurate predictor for near fields and far fields of arbitrary 3D nanostructures
Deep learning meets nanophotonics: A generalized Accurate predictor for near fields and far fields of arbitrary 3D nanostructures
Deep artificial neural networks are powerful tools with many possible applications in nanophotonics. Here, we demonstrate how a deep neural network can be used as a fast, general purpose predictor of the full near-field and far-field response of plasmonic and dielectric nanostructures. A trained neural network is shown to infer the internal fields of arbitrary three-dimensional nanostructures many orders of magnitude faster compared to conventional numerical simulations. Secondary physical quantities are derived from the deep learning predictions and faithfully reproduce a wide variety of physical effects without requiring specific training. We discuss the strengths and limitations of the neural network approach using a number of model studies of single particles and their near-field interactions. Our approach paves the way for fast, yet universal, methods for design and analysis of nanophotonic systems.
Deep learning, nanophotonics, plasmonics, rapid nano-optics simulations, silicon nanostructures
329-338
Wiecha, Peter R.
fb335482-9577-41af-a0ef-3988b7654c9b
Muskens, Otto L.
2284101a-f9ef-4d79-8951-a6cda5bfc7f9
8 January 2020
Wiecha, Peter R.
fb335482-9577-41af-a0ef-3988b7654c9b
Muskens, Otto L.
2284101a-f9ef-4d79-8951-a6cda5bfc7f9
Wiecha, Peter R. and Muskens, Otto L.
(2020)
Deep learning meets nanophotonics: A generalized Accurate predictor for near fields and far fields of arbitrary 3D nanostructures.
Nano Letters, 20 (1), .
(doi:10.1021/acs.nanolett.9b03971).
Abstract
Deep artificial neural networks are powerful tools with many possible applications in nanophotonics. Here, we demonstrate how a deep neural network can be used as a fast, general purpose predictor of the full near-field and far-field response of plasmonic and dielectric nanostructures. A trained neural network is shown to infer the internal fields of arbitrary three-dimensional nanostructures many orders of magnitude faster compared to conventional numerical simulations. Secondary physical quantities are derived from the deep learning predictions and faithfully reproduce a wide variety of physical effects without requiring specific training. We discuss the strengths and limitations of the neural network approach using a number of model studies of single particles and their near-field interactions. Our approach paves the way for fast, yet universal, methods for design and analysis of nanophotonic systems.
Text
WiechaNanoLett2020
- Accepted Manuscript
More information
e-pub ahead of print date: 11 December 2019
Published date: 8 January 2020
Keywords:
Deep learning, nanophotonics, plasmonics, rapid nano-optics simulations, silicon nanostructures
Identifiers
Local EPrints ID: 437216
URI: http://eprints.soton.ac.uk/id/eprint/437216
ISSN: 1530-6984
PURE UUID: 9e907ccc-eaf3-4ba3-ae47-94ac469bf1f0
Catalogue record
Date deposited: 22 Jan 2020 17:31
Last modified: 17 Mar 2024 05:13
Export record
Altmetrics
Contributors
Author:
Peter R. Wiecha
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
