Knowledge Discovery in Nanophotonics Using Geometric Deep Learning
Herein, a new approach for using the intelligence aspects of artificial intelligence for knowledge discovery rather than device optimization in electromagnetic (EM) nanostructures is presented. This approach uses training data obtained through full‐wave EM simulations of a series of nanostructures t...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-02-01
|
| Series: | Advanced Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/aisy.201900132 |
| _version_ | 1811324375330193408 |
|---|---|
| author | Yashar Kiarashinejad Mohammadreza Zandehshahvar Sajjad Abdollahramezani Omid Hemmatyar Reza Pourabolghasem Ali Adibi |
| author_facet | Yashar Kiarashinejad Mohammadreza Zandehshahvar Sajjad Abdollahramezani Omid Hemmatyar Reza Pourabolghasem Ali Adibi |
| author_sort | Yashar Kiarashinejad |
| collection | DOAJ |
| description | Herein, a new approach for using the intelligence aspects of artificial intelligence for knowledge discovery rather than device optimization in electromagnetic (EM) nanostructures is presented. This approach uses training data obtained through full‐wave EM simulations of a series of nanostructures to train geometric deep learning algorithms to assess the range of feasible responses as well as the feasibility of a desired response from a class of EM nanostructures. To facilitate the knowledge discovery, this approach combines the dimensionality reduction technique with convex‐hull and one‐class support‐vector‐machine (SVM) algorithms to find the range of the feasible responses in the latent response space of the EM nanostructure. More importantly, the one‐class SVM algorithm can be trained to provide the degree of feasibility of a response from a given nanostructure. This important information can be used to modify the initial structure to an alternative one that can enable an initially unfeasible response. To show the applicability of this approach, it is applied to two important classes of binary metasurfaces (MSs), formed by an array of plasmonic nanostructures, and periodic MSs formed by an array of dielectric nanopillars. These theoretical and experimental results confirm the unique features of this approach for knowledge discovery in EM nanostructures. |
| first_indexed | 2024-04-13T14:13:08Z |
| format | Article |
| id | doaj.art-5df577fa933e4462a307ffb2f91d0eb1 |
| institution | Directory Open Access Journal |
| issn | 2640-4567 |
| language | English |
| last_indexed | 2024-04-13T14:13:08Z |
| publishDate | 2020-02-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advanced Intelligent Systems |
| spelling | doaj.art-5df577fa933e4462a307ffb2f91d0eb12022-12-22T02:43:43ZengWileyAdvanced Intelligent Systems2640-45672020-02-0122n/an/a10.1002/aisy.201900132Knowledge Discovery in Nanophotonics Using Geometric Deep LearningYashar Kiarashinejad0Mohammadreza Zandehshahvar1Sajjad Abdollahramezani2Omid Hemmatyar3Reza Pourabolghasem4Ali Adibi5Georgia Institute of Technology 778 Atlantic Drive NW Atlanta GA 30332 USAGeorgia Institute of Technology 778 Atlantic Drive NW Atlanta GA 30332 USAGeorgia Institute of Technology 778 Atlantic Drive NW Atlanta GA 30332 USAGeorgia Institute of Technology 778 Atlantic Drive NW Atlanta GA 30332 USAGeorgia Institute of Technology 778 Atlantic Drive NW Atlanta GA 30332 USAGeorgia Institute of Technology 778 Atlantic Drive NW Atlanta GA 30332 USAHerein, a new approach for using the intelligence aspects of artificial intelligence for knowledge discovery rather than device optimization in electromagnetic (EM) nanostructures is presented. This approach uses training data obtained through full‐wave EM simulations of a series of nanostructures to train geometric deep learning algorithms to assess the range of feasible responses as well as the feasibility of a desired response from a class of EM nanostructures. To facilitate the knowledge discovery, this approach combines the dimensionality reduction technique with convex‐hull and one‐class support‐vector‐machine (SVM) algorithms to find the range of the feasible responses in the latent response space of the EM nanostructure. More importantly, the one‐class SVM algorithm can be trained to provide the degree of feasibility of a response from a given nanostructure. This important information can be used to modify the initial structure to an alternative one that can enable an initially unfeasible response. To show the applicability of this approach, it is applied to two important classes of binary metasurfaces (MSs), formed by an array of plasmonic nanostructures, and periodic MSs formed by an array of dielectric nanopillars. These theoretical and experimental results confirm the unique features of this approach for knowledge discovery in EM nanostructures.https://doi.org/10.1002/aisy.201900132artificial intelligencedeep learninglight–matter interactionsmachine learningnanophotonics |
| spellingShingle | Yashar Kiarashinejad Mohammadreza Zandehshahvar Sajjad Abdollahramezani Omid Hemmatyar Reza Pourabolghasem Ali Adibi Knowledge Discovery in Nanophotonics Using Geometric Deep Learning Advanced Intelligent Systems artificial intelligence deep learning light–matter interactions machine learning nanophotonics |
| title | Knowledge Discovery in Nanophotonics Using Geometric Deep Learning |
| title_full | Knowledge Discovery in Nanophotonics Using Geometric Deep Learning |
| title_fullStr | Knowledge Discovery in Nanophotonics Using Geometric Deep Learning |
| title_full_unstemmed | Knowledge Discovery in Nanophotonics Using Geometric Deep Learning |
| title_short | Knowledge Discovery in Nanophotonics Using Geometric Deep Learning |
| title_sort | knowledge discovery in nanophotonics using geometric deep learning |
| topic | artificial intelligence deep learning light–matter interactions machine learning nanophotonics |
| url | https://doi.org/10.1002/aisy.201900132 |
| work_keys_str_mv | AT yasharkiarashinejad knowledgediscoveryinnanophotonicsusinggeometricdeeplearning AT mohammadrezazandehshahvar knowledgediscoveryinnanophotonicsusinggeometricdeeplearning AT sajjadabdollahramezani knowledgediscoveryinnanophotonicsusinggeometricdeeplearning AT omidhemmatyar knowledgediscoveryinnanophotonicsusinggeometricdeeplearning AT rezapourabolghasem knowledgediscoveryinnanophotonicsusinggeometricdeeplearning AT aliadibi knowledgediscoveryinnanophotonicsusinggeometricdeeplearning |