Deep empirical neural network for optical phase retrieval over a scattering medium
Abstract Supervised learning, a popular tool in modern science and technology, thrives on huge amounts of labeled data. Physics-enhanced deep neural networks offer an effective solution to alleviate the data burden by incorporating an analytical model that interprets the underlying physical processe...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-02-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-56522-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861814462513152 |
|---|---|
| author | Huaisheng Tu Haotian Liu Tuqiang Pan Wuping Xie Zihao Ma Fan Zhang Pengbai Xu Leiming Wu Ou Xu Yi Xu Yuwen Qin |
| author_facet | Huaisheng Tu Haotian Liu Tuqiang Pan Wuping Xie Zihao Ma Fan Zhang Pengbai Xu Leiming Wu Ou Xu Yi Xu Yuwen Qin |
| author_sort | Huaisheng Tu |
| collection | DOAJ |
| description | Abstract Supervised learning, a popular tool in modern science and technology, thrives on huge amounts of labeled data. Physics-enhanced deep neural networks offer an effective solution to alleviate the data burden by incorporating an analytical model that interprets the underlying physical processes. However, it completely fails in tackling systems without analytical solution, where wave scattering systems with multiple input multiple output are typical examples. Herein, we propose a concept of deep empirical neural network (DENN) that is a hybridization of a deep neural network and an empirical model, which enables seeing through an opaque scattering medium in an untrained manner. The DENN does not rely on labeled data, all while delivering as high as 58% improvement in fidelity compared with the supervised learning using 30000 data pairs for achieving the same goal of optical phase retrieval. The DENN might shed new light on the applications of deep learning in physics, information science, biology, chemistry and beyond. |
| format | Article |
| id | doaj-art-747f7a7b3fad4ba9bde0923c8bcf65fe |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-747f7a7b3fad4ba9bde0923c8bcf65fe2025-02-09T12:45:47ZengNature PortfolioNature Communications2041-17232025-02-011611910.1038/s41467-025-56522-5Deep empirical neural network for optical phase retrieval over a scattering mediumHuaisheng Tu0Haotian Liu1Tuqiang Pan2Wuping Xie3Zihao Ma4Fan Zhang5Pengbai Xu6Leiming Wu7Ou Xu8Yi Xu9Yuwen Qin10Key Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyKey Laboratory of Photonic Technology for Integrated Sensing and Communication, Ministry of Education, Guangdong University of TechnologyAbstract Supervised learning, a popular tool in modern science and technology, thrives on huge amounts of labeled data. Physics-enhanced deep neural networks offer an effective solution to alleviate the data burden by incorporating an analytical model that interprets the underlying physical processes. However, it completely fails in tackling systems without analytical solution, where wave scattering systems with multiple input multiple output are typical examples. Herein, we propose a concept of deep empirical neural network (DENN) that is a hybridization of a deep neural network and an empirical model, which enables seeing through an opaque scattering medium in an untrained manner. The DENN does not rely on labeled data, all while delivering as high as 58% improvement in fidelity compared with the supervised learning using 30000 data pairs for achieving the same goal of optical phase retrieval. The DENN might shed new light on the applications of deep learning in physics, information science, biology, chemistry and beyond.https://doi.org/10.1038/s41467-025-56522-5 |
| spellingShingle | Huaisheng Tu Haotian Liu Tuqiang Pan Wuping Xie Zihao Ma Fan Zhang Pengbai Xu Leiming Wu Ou Xu Yi Xu Yuwen Qin Deep empirical neural network for optical phase retrieval over a scattering medium Nature Communications |
| title | Deep empirical neural network for optical phase retrieval over a scattering medium |
| title_full | Deep empirical neural network for optical phase retrieval over a scattering medium |
| title_fullStr | Deep empirical neural network for optical phase retrieval over a scattering medium |
| title_full_unstemmed | Deep empirical neural network for optical phase retrieval over a scattering medium |
| title_short | Deep empirical neural network for optical phase retrieval over a scattering medium |
| title_sort | deep empirical neural network for optical phase retrieval over a scattering medium |
| url | https://doi.org/10.1038/s41467-025-56522-5 |
| work_keys_str_mv | AT huaishengtu deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT haotianliu deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT tuqiangpan deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT wupingxie deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT zihaoma deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT fanzhang deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT pengbaixu deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT leimingwu deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT ouxu deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT yixu deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium AT yuwenqin deepempiricalneuralnetworkforopticalphaseretrievaloverascatteringmedium |