Transfer Learning for Modeling Plasmonic Nanowire Waveguides
Retrieving waveguiding properties of plasmonic metal nanowires (MNWs) through numerical simulations is time- and computational-resource-consuming, especially for those with abrupt geometric features and broken symmetries. Deep learning provides an alternative approach but is challenging to use due t...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Nanomaterials |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-4991/12/20/3624 |
| _version_ | 1827648552761819136 |
|---|---|
| author | Aoning Luo Yuanjia Feng Chunyan Zhu Yipei Wang Xiaoqin Wu |
| author_facet | Aoning Luo Yuanjia Feng Chunyan Zhu Yipei Wang Xiaoqin Wu |
| author_sort | Aoning Luo |
| collection | DOAJ |
| description | Retrieving waveguiding properties of plasmonic metal nanowires (MNWs) through numerical simulations is time- and computational-resource-consuming, especially for those with abrupt geometric features and broken symmetries. Deep learning provides an alternative approach but is challenging to use due to inadequate generalization performance and the requirement of large sets of training data. Here, we overcome these constraints by proposing a transfer learning approach for modeling MNWs under the guidance of physics. We show that the basic knowledge of plasmon modes can first be learned from free-standing circular MNWs with computationally inexpensive data, and then reused to significantly improve performance in predicting waveguiding properties of MNWs with various complex configurations, enabling much smaller errors (~23–61% reduction), less trainable parameters (~42% reduction), and smaller sets of training data (~50–80% reduction) than direct learning. Compared to numerical simulations, our model reduces the computational time by five orders of magnitude. Compared to other non-deep learning methods, such as the circular-area-equivalence approach and the diagonal-circle approximation, our approach enables not only much higher accuracies, but also more comprehensive characterizations, offering an effective and efficient framework to investigate MNWs that may greatly facilitate the design of polaritonic components and devices. |
| first_indexed | 2024-03-09T19:40:49Z |
| format | Article |
| id | doaj.art-d1861a79ed52419b8e41b9dc16316d15 |
| institution | Directory Open Access Journal |
| issn | 2079-4991 |
| language | English |
| last_indexed | 2024-03-09T19:40:49Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Nanomaterials |
| spelling | doaj.art-d1861a79ed52419b8e41b9dc16316d152023-11-24T01:40:30ZengMDPI AGNanomaterials2079-49912022-10-011220362410.3390/nano12203624Transfer Learning for Modeling Plasmonic Nanowire WaveguidesAoning Luo0Yuanjia Feng1Chunyan Zhu2Yipei Wang3Xiaoqin Wu4Key Laboratory of Optoelectronic Technology and Systems (Ministry of Education), College of Optoelectronic Engineering, Chongqing University, Chongqing 400044, ChinaKey Laboratory of Optoelectronic Technology and Systems (Ministry of Education), College of Optoelectronic Engineering, Chongqing University, Chongqing 400044, ChinaKey Laboratory of Optoelectronic Technology and Systems (Ministry of Education), College of Optoelectronic Engineering, Chongqing University, Chongqing 400044, ChinaKey Laboratory of Optoelectronic Technology and Systems (Ministry of Education), College of Optoelectronic Engineering, Chongqing University, Chongqing 400044, ChinaKey Laboratory of Optoelectronic Technology and Systems (Ministry of Education), College of Optoelectronic Engineering, Chongqing University, Chongqing 400044, ChinaRetrieving waveguiding properties of plasmonic metal nanowires (MNWs) through numerical simulations is time- and computational-resource-consuming, especially for those with abrupt geometric features and broken symmetries. Deep learning provides an alternative approach but is challenging to use due to inadequate generalization performance and the requirement of large sets of training data. Here, we overcome these constraints by proposing a transfer learning approach for modeling MNWs under the guidance of physics. We show that the basic knowledge of plasmon modes can first be learned from free-standing circular MNWs with computationally inexpensive data, and then reused to significantly improve performance in predicting waveguiding properties of MNWs with various complex configurations, enabling much smaller errors (~23–61% reduction), less trainable parameters (~42% reduction), and smaller sets of training data (~50–80% reduction) than direct learning. Compared to numerical simulations, our model reduces the computational time by five orders of magnitude. Compared to other non-deep learning methods, such as the circular-area-equivalence approach and the diagonal-circle approximation, our approach enables not only much higher accuracies, but also more comprehensive characterizations, offering an effective and efficient framework to investigate MNWs that may greatly facilitate the design of polaritonic components and devices.https://www.mdpi.com/2079-4991/12/20/3624deep learningtransfer learningplasmonicsnanowireswaveguides |
| spellingShingle | Aoning Luo Yuanjia Feng Chunyan Zhu Yipei Wang Xiaoqin Wu Transfer Learning for Modeling Plasmonic Nanowire Waveguides Nanomaterials deep learning transfer learning plasmonics nanowires waveguides |
| title | Transfer Learning for Modeling Plasmonic Nanowire Waveguides |
| title_full | Transfer Learning for Modeling Plasmonic Nanowire Waveguides |
| title_fullStr | Transfer Learning for Modeling Plasmonic Nanowire Waveguides |
| title_full_unstemmed | Transfer Learning for Modeling Plasmonic Nanowire Waveguides |
| title_short | Transfer Learning for Modeling Plasmonic Nanowire Waveguides |
| title_sort | transfer learning for modeling plasmonic nanowire waveguides |
| topic | deep learning transfer learning plasmonics nanowires waveguides |
| url | https://www.mdpi.com/2079-4991/12/20/3624 |
| work_keys_str_mv | AT aoningluo transferlearningformodelingplasmonicnanowirewaveguides AT yuanjiafeng transferlearningformodelingplasmonicnanowirewaveguides AT chunyanzhu transferlearningformodelingplasmonicnanowirewaveguides AT yipeiwang transferlearningformodelingplasmonicnanowirewaveguides AT xiaoqinwu transferlearningformodelingplasmonicnanowirewaveguides |