A knowledge-inherited learning for intelligent metasurface design and assembly
Abstract Recent breakthroughs in deep learning have ushered in an essential tool for optics and photonics, recurring in various applications of material design, system optimization, and automation control. Deep learning-enabled on-demand metasurface design has been the subject of extensive expansion...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Publishing Group
2023-03-01
|
| Series: | Light: Science & Applications |
| Online Access: | https://doi.org/10.1038/s41377-023-01131-4 |
| _version_ | 1797853578350559232 |
|---|---|
| author | Yuetian Jia Chao Qian Zhixiang Fan Tong Cai Er-Ping Li Hongsheng Chen |
| author_facet | Yuetian Jia Chao Qian Zhixiang Fan Tong Cai Er-Ping Li Hongsheng Chen |
| author_sort | Yuetian Jia |
| collection | DOAJ |
| description | Abstract Recent breakthroughs in deep learning have ushered in an essential tool for optics and photonics, recurring in various applications of material design, system optimization, and automation control. Deep learning-enabled on-demand metasurface design has been the subject of extensive expansion, as it can alleviate the time-consuming, low-efficiency, and experience-orientated shortcomings in conventional numerical simulations and physics-based methods. However, collecting samples and training neural networks are fundamentally confined to predefined individual metamaterials and tend to fail for large problem sizes. Inspired by object-oriented C++ programming, we propose a knowledge-inherited paradigm for multi-object and shape-unbound metasurface inverse design. Each inherited neural network carries knowledge from the “parent” metasurface and then is freely assembled to construct the “offspring” metasurface; such a process is as simple as building a container-type house. We benchmark the paradigm by the free design of aperiodic and periodic metasurfaces, with accuracies that reach 86.7%. Furthermore, we present an intelligent origami metasurface to facilitate compatible and lightweight satellite communication facilities. Our work opens up a new avenue for automatic metasurface design and leverages the assemblability to broaden the adaptability of intelligent metadevices. |
| first_indexed | 2024-04-09T19:51:47Z |
| format | Article |
| id | doaj.art-4d73503bfeca4bdbbb2fdb264e486f13 |
| institution | Directory Open Access Journal |
| issn | 2047-7538 |
| language | English |
| last_indexed | 2024-04-09T19:51:47Z |
| publishDate | 2023-03-01 |
| publisher | Nature Publishing Group |
| record_format | Article |
| series | Light: Science & Applications |
| spelling | doaj.art-4d73503bfeca4bdbbb2fdb264e486f132023-04-03T05:41:17ZengNature Publishing GroupLight: Science & Applications2047-75382023-03-0112111110.1038/s41377-023-01131-4A knowledge-inherited learning for intelligent metasurface design and assemblyYuetian Jia0Chao Qian1Zhixiang Fan2Tong Cai3Er-Ping Li4Hongsheng Chen5ZJU-UIUC Institute, Interdisciplinary Center for Quantum Information, State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang UniversityZJU-UIUC Institute, Interdisciplinary Center for Quantum Information, State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang UniversityZJU-UIUC Institute, Interdisciplinary Center for Quantum Information, State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang UniversityZJU-UIUC Institute, Interdisciplinary Center for Quantum Information, State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang UniversityZJU-UIUC Institute, Interdisciplinary Center for Quantum Information, State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang UniversityZJU-UIUC Institute, Interdisciplinary Center for Quantum Information, State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang UniversityAbstract Recent breakthroughs in deep learning have ushered in an essential tool for optics and photonics, recurring in various applications of material design, system optimization, and automation control. Deep learning-enabled on-demand metasurface design has been the subject of extensive expansion, as it can alleviate the time-consuming, low-efficiency, and experience-orientated shortcomings in conventional numerical simulations and physics-based methods. However, collecting samples and training neural networks are fundamentally confined to predefined individual metamaterials and tend to fail for large problem sizes. Inspired by object-oriented C++ programming, we propose a knowledge-inherited paradigm for multi-object and shape-unbound metasurface inverse design. Each inherited neural network carries knowledge from the “parent” metasurface and then is freely assembled to construct the “offspring” metasurface; such a process is as simple as building a container-type house. We benchmark the paradigm by the free design of aperiodic and periodic metasurfaces, with accuracies that reach 86.7%. Furthermore, we present an intelligent origami metasurface to facilitate compatible and lightweight satellite communication facilities. Our work opens up a new avenue for automatic metasurface design and leverages the assemblability to broaden the adaptability of intelligent metadevices.https://doi.org/10.1038/s41377-023-01131-4 |
| spellingShingle | Yuetian Jia Chao Qian Zhixiang Fan Tong Cai Er-Ping Li Hongsheng Chen A knowledge-inherited learning for intelligent metasurface design and assembly Light: Science & Applications |
| title | A knowledge-inherited learning for intelligent metasurface design and assembly |
| title_full | A knowledge-inherited learning for intelligent metasurface design and assembly |
| title_fullStr | A knowledge-inherited learning for intelligent metasurface design and assembly |
| title_full_unstemmed | A knowledge-inherited learning for intelligent metasurface design and assembly |
| title_short | A knowledge-inherited learning for intelligent metasurface design and assembly |
| title_sort | knowledge inherited learning for intelligent metasurface design and assembly |
| url | https://doi.org/10.1038/s41377-023-01131-4 |
| work_keys_str_mv | AT yuetianjia aknowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT chaoqian aknowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT zhixiangfan aknowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT tongcai aknowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT erpingli aknowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT hongshengchen aknowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT yuetianjia knowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT chaoqian knowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT zhixiangfan knowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT tongcai knowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT erpingli knowledgeinheritedlearningforintelligentmetasurfacedesignandassembly AT hongshengchen knowledgeinheritedlearningforintelligentmetasurfacedesignandassembly |