POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal Nanocavities
We study a new technique for solving the fundamental challenge in nanophotonic design: fast and accurate characterization of nanoscale photonic devices with minimal human intervention. Much like the fusion between Artificial Intelligence and Electronic Design Automation (EDA), many efforts have been...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Nanomaterials |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-4991/12/24/4401 |
| _version_ | 1797455919536144384 |
|---|---|
| author | Xinyu Chen Renjie Li Yueyao Yu Yuanwen Shen Wenye Li Yin Zhang Zhaoyu Zhang |
| author_facet | Xinyu Chen Renjie Li Yueyao Yu Yuanwen Shen Wenye Li Yin Zhang Zhaoyu Zhang |
| author_sort | Xinyu Chen |
| collection | DOAJ |
| description | We study a new technique for solving the fundamental challenge in nanophotonic design: fast and accurate characterization of nanoscale photonic devices with minimal human intervention. Much like the fusion between Artificial Intelligence and Electronic Design Automation (EDA), many efforts have been made to apply deep neural networks (DNN) such as convolutional neural networks to prototype and characterize next-gen optoelectronic devices commonly found in Photonic Integrated Circuits. However, state-of-the-art DNN models are still far from being directly applicable in the real world: e.g., DNN-produced correlation coefficients between target and predicted physical quantities are about 80%, which is much lower than what it takes to generate reliable and reproducible nanophotonic designs. Recently, attention-based transformer models have attracted extensive interests and been widely used in Computer Vision and Natural Language Processing. In this work, we for the first time propose a Transformer model (POViT) to efficiently design and simulate photonic crystal nanocavities with multiple objectives under consideration. Unlike the standard Vision Transformer, our model takes photonic crystals as input data and changes the activation layer from GELU to an absolute-value function. Extensive experiments show that POViT significantly improves results reported by previous models: correlation coefficients are increased by over <inline-formula><math display="inline"><semantics><mrow><mn>12</mn><mo>%</mo></mrow></semantics></math></inline-formula> (i.e., to <inline-formula><math display="inline"><semantics><mrow><mn>92.0</mn><mo>%</mo></mrow></semantics></math></inline-formula>) and prediction errors are reduced by an order of magnitude, among several key metric improvements. Our work has the potential to drive the expansion of EDA to fully automated photonic design (i.e., PDA). The complete dataset and code will be released to promote research in the interdisciplinary field of materials science/physics and computer science. |
| first_indexed | 2024-03-09T16:01:08Z |
| format | Article |
| id | doaj.art-3933176778b14b36ac5af7bceae5a0f3 |
| institution | Directory Open Access Journal |
| issn | 2079-4991 |
| language | English |
| last_indexed | 2024-03-09T16:01:08Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Nanomaterials |
| spelling | doaj.art-3933176778b14b36ac5af7bceae5a0f32023-11-24T17:03:48ZengMDPI AGNanomaterials2079-49912022-12-011224440110.3390/nano12244401POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal NanocavitiesXinyu Chen0Renjie Li1Yueyao Yu2Yuanwen Shen3Wenye Li4Yin Zhang5Zhaoyu Zhang6Shenzhen Key Laboratory of Semiconductor Lasers, School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, 2001 Longxiang Ave, Shenzhen 518172, ChinaShenzhen Key Laboratory of Semiconductor Lasers, School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, 2001 Longxiang Ave, Shenzhen 518172, ChinaShenzhen Research Institute of Big Data (SRIBD), 2001 Longxiang Ave, Shenzhen 518172, ChinaShenzhen Key Laboratory of Semiconductor Lasers, School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, 2001 Longxiang Ave, Shenzhen 518172, ChinaShenzhen Research Institute of Big Data (SRIBD), 2001 Longxiang Ave, Shenzhen 518172, ChinaSchool of Data Science, The Chinese University of Hong Kong, Shenzhen, 2001 Longxiang Ave, Shenzhen 518172, ChinaShenzhen Key Laboratory of Semiconductor Lasers, School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, 2001 Longxiang Ave, Shenzhen 518172, ChinaWe study a new technique for solving the fundamental challenge in nanophotonic design: fast and accurate characterization of nanoscale photonic devices with minimal human intervention. Much like the fusion between Artificial Intelligence and Electronic Design Automation (EDA), many efforts have been made to apply deep neural networks (DNN) such as convolutional neural networks to prototype and characterize next-gen optoelectronic devices commonly found in Photonic Integrated Circuits. However, state-of-the-art DNN models are still far from being directly applicable in the real world: e.g., DNN-produced correlation coefficients between target and predicted physical quantities are about 80%, which is much lower than what it takes to generate reliable and reproducible nanophotonic designs. Recently, attention-based transformer models have attracted extensive interests and been widely used in Computer Vision and Natural Language Processing. In this work, we for the first time propose a Transformer model (POViT) to efficiently design and simulate photonic crystal nanocavities with multiple objectives under consideration. Unlike the standard Vision Transformer, our model takes photonic crystals as input data and changes the activation layer from GELU to an absolute-value function. Extensive experiments show that POViT significantly improves results reported by previous models: correlation coefficients are increased by over <inline-formula><math display="inline"><semantics><mrow><mn>12</mn><mo>%</mo></mrow></semantics></math></inline-formula> (i.e., to <inline-formula><math display="inline"><semantics><mrow><mn>92.0</mn><mo>%</mo></mrow></semantics></math></inline-formula>) and prediction errors are reduced by an order of magnitude, among several key metric improvements. Our work has the potential to drive the expansion of EDA to fully automated photonic design (i.e., PDA). The complete dataset and code will be released to promote research in the interdisciplinary field of materials science/physics and computer science.https://www.mdpi.com/2079-4991/12/24/4401deep learningvision transformernanophotonicslasersphotonic crystals |
| spellingShingle | Xinyu Chen Renjie Li Yueyao Yu Yuanwen Shen Wenye Li Yin Zhang Zhaoyu Zhang POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal Nanocavities Nanomaterials deep learning vision transformer nanophotonics lasers photonic crystals |
| title | POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal Nanocavities |
| title_full | POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal Nanocavities |
| title_fullStr | POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal Nanocavities |
| title_full_unstemmed | POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal Nanocavities |
| title_short | POViT: Vision Transformer for Multi-Objective Design and Characterization of Photonic Crystal Nanocavities |
| title_sort | povit vision transformer for multi objective design and characterization of photonic crystal nanocavities |
| topic | deep learning vision transformer nanophotonics lasers photonic crystals |
| url | https://www.mdpi.com/2079-4991/12/24/4401 |
| work_keys_str_mv | AT xinyuchen povitvisiontransformerformultiobjectivedesignandcharacterizationofphotoniccrystalnanocavities AT renjieli povitvisiontransformerformultiobjectivedesignandcharacterizationofphotoniccrystalnanocavities AT yueyaoyu povitvisiontransformerformultiobjectivedesignandcharacterizationofphotoniccrystalnanocavities AT yuanwenshen povitvisiontransformerformultiobjectivedesignandcharacterizationofphotoniccrystalnanocavities AT wenyeli povitvisiontransformerformultiobjectivedesignandcharacterizationofphotoniccrystalnanocavities AT yinzhang povitvisiontransformerformultiobjectivedesignandcharacterizationofphotoniccrystalnanocavities AT zhaoyuzhang povitvisiontransformerformultiobjectivedesignandcharacterizationofphotoniccrystalnanocavities |