Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networks
Summary: A hollow triangular rod-type valley-Hall photonic topological insulator is proposed, and two tandem residual deep neural networks are built for multimodal inverse design of the structure. One of them is a tandem multilayer perceptron, and the other is a composite tandem network based on var...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-04-01
|
| Series: | iScience |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2589004225005371 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849763905694334976 |
|---|---|
| author | Bing-Jiang Wang Le Zhang Ben-Xin Wang Dong-Ping Zhang Ya-Guang Xie Jin-Hui Cai |
| author_facet | Bing-Jiang Wang Le Zhang Ben-Xin Wang Dong-Ping Zhang Ya-Guang Xie Jin-Hui Cai |
| author_sort | Bing-Jiang Wang |
| collection | DOAJ |
| description | Summary: A hollow triangular rod-type valley-Hall photonic topological insulator is proposed, and two tandem residual deep neural networks are built for multimodal inverse design of the structure. One of them is a tandem multilayer perceptron, and the other is a composite tandem network based on variational auto-encoder. The former is used to inversely infer the value of the structural sizes, and the latter is used to predict the structural image of the lattice from demanded design targets. Residual connections are included in both networks to speed up the training convergence as well as avoid vanishing gradient problem. Based on an arbitrary inversely designed lattice, domain walls between two photonic topological insulators with different topology are constructed, and full-wave simulations on the transmission properties are conducted. Numerical results show that robust topologically protected wave propagation is supported along the domain wall with little backscattering, demonstrating that the proposed methods are valid. |
| format | Article |
| id | doaj-art-a13cf04fbd2840baae744a4b6c76a062 |
| institution | DOAJ |
| issn | 2589-0042 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | iScience |
| spelling | doaj-art-a13cf04fbd2840baae744a4b6c76a0622025-08-20T03:05:17ZengElsevieriScience2589-00422025-04-0128411227610.1016/j.isci.2025.112276Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networksBing-Jiang Wang0Le Zhang1Ben-Xin Wang2Dong-Ping Zhang3Ya-Guang Xie4Jin-Hui Cai5Centre for THz Research, College of Information Engineering, China Jiliang University, Hangzhou 310018, ChinaCentre for THz Research, College of Information Engineering, China Jiliang University, Hangzhou 310018, China; Corresponding authorSchool of Science, Jiangnan University, Wuxi 214122, China; Corresponding authorCentre for THz Research, College of Information Engineering, China Jiliang University, Hangzhou 310018, ChinaHangzhou Arcvideo Technology Co.,Ltd., Hangzhou 310051, ChinaCollege of Metrology & Measurement Engineering, China Jiliang University, Hangzhou 310018, ChinaSummary: A hollow triangular rod-type valley-Hall photonic topological insulator is proposed, and two tandem residual deep neural networks are built for multimodal inverse design of the structure. One of them is a tandem multilayer perceptron, and the other is a composite tandem network based on variational auto-encoder. The former is used to inversely infer the value of the structural sizes, and the latter is used to predict the structural image of the lattice from demanded design targets. Residual connections are included in both networks to speed up the training convergence as well as avoid vanishing gradient problem. Based on an arbitrary inversely designed lattice, domain walls between two photonic topological insulators with different topology are constructed, and full-wave simulations on the transmission properties are conducted. Numerical results show that robust topologically protected wave propagation is supported along the domain wall with little backscattering, demonstrating that the proposed methods are valid.http://www.sciencedirect.com/science/article/pii/S2589004225005371Topological photonicsDeep learningInverse design |
| spellingShingle | Bing-Jiang Wang Le Zhang Ben-Xin Wang Dong-Ping Zhang Ya-Guang Xie Jin-Hui Cai Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networks iScience Topological photonics Deep learning Inverse design |
| title | Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networks |
| title_full | Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networks |
| title_fullStr | Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networks |
| title_full_unstemmed | Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networks |
| title_short | Inverse design of a valley-Hall photonic topological insulator based on tandem residual neural networks |
| title_sort | inverse design of a valley hall photonic topological insulator based on tandem residual neural networks |
| topic | Topological photonics Deep learning Inverse design |
| url | http://www.sciencedirect.com/science/article/pii/S2589004225005371 |
| work_keys_str_mv | AT bingjiangwang inversedesignofavalleyhallphotonictopologicalinsulatorbasedontandemresidualneuralnetworks AT lezhang inversedesignofavalleyhallphotonictopologicalinsulatorbasedontandemresidualneuralnetworks AT benxinwang inversedesignofavalleyhallphotonictopologicalinsulatorbasedontandemresidualneuralnetworks AT dongpingzhang inversedesignofavalleyhallphotonictopologicalinsulatorbasedontandemresidualneuralnetworks AT yaguangxie inversedesignofavalleyhallphotonictopologicalinsulatorbasedontandemresidualneuralnetworks AT jinhuicai inversedesignofavalleyhallphotonictopologicalinsulatorbasedontandemresidualneuralnetworks |