journal article·2024-09Open access

Harnessing the Missing Spectral Correlation for Metasurface Inverse Design

Jie Zhang, Chao Qian, Guangfeng You, Tao Wang, Yasir Saifullah, Reza Abdi‐Ghaleh, Hongsheng Chen

Advanced Science· Volume 11· Issue 33

Abstract

Abstract A long‐held tenet in computer science asserts that the training of deep learning is analogous to an alchemical furnace, and its “black box” signature brings forth inexplicability. For electromagnetic metasurfaces, the related intelligent applications also get stuck into such a dilemma. Although the past 5 years have witnessed a proliferation of deep learning‐based works across complex photonic scenarios, they neglect the already existing but untapped physical laws. Here, the intrinsic correlation between the real and imaginary parts of the spectra are revealed using Kramers–Kronig relations, which is then mimicked by bidirectional information flow in neural network space. Such consideration harnesses the missing spectral connection to extract crucial features effectively. The bidirectional recurrent neural network is benchmarked in metasurface inverse design and compare it with a fully‐connected neural network, unidirectional recurrent neural network, and attention‐based transformer. Beyond the improved accuracy, the study examines the intermediate information products and physically explains why different network structures yield different performances. The work offers explicable perspectives to utilize physical information in the deep learning field and facilitates many data‐intensive research endeavors.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Metamaterials and Metasurfaces Applications · Advanced Antenna and Metasurface Technologies · Advanced Optical Imaging Technologies

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

ZJU‐UIUC Institute, Interdisciplinary Center for Quantum Information State Key Laboratory of Extreme Photonics and Instrumentation Zhejiang University Hangzhou 310027 China

ZJU‐Hangzhou Global Science and Technology Innovation Center Key Lab. of Advanced Micro/Nano Electronic Devices & Smart Systems of Zhejiang Zhejiang University Hangzhou 310027 China

Jinhua Institute of Zhejiang University Zhejiang University Jinhua 321099 China

State Key Laboratory of Integrated Service Networks Xidian University Xian 710071 China

Department of Laser and Optical Engineering University of Bonab Bonab 5551395133 Iran

Funding

National Natural Science Foundation of China · Fundamental Research Funds for the Central Universities

References (52)

Show source-provided references
1. 10.1038/nmat383910.1038/nmat3839 ↗
2. 10.1038/natrevmats.2016.6710.1038/natrevmats.2016.67 ↗
3. 10.1038/s41467-023-41123-x10.1038/s41467-023-41123-x ↗
4. 10.1038/s41377-021-00507-810.1038/s41377-021-00507-8 ↗
5. 10.1038/s41566-020-0604-210.1038/s41566-020-0604-2 ↗
6. 10.1002/adma.20220505310.1002/adma.202205053 ↗
7. 10.1038/s41928-020-0380-510.1038/s41928-020-0380-5 ↗
8. 10.1364/PRJ.41844510.1364/PRJ.418445 ↗
9. 10.1002/adfm.20210933110.1002/adfm.202109331 ↗
10. 10.1103/PhysRevB.91.19544410.1103/PhysRevB.91.195444 ↗
11. Qian C. · Appl. Phys. Lett. · 2021 · 118 · 18
12. 10.1002/adom.20220213010.1002/adom.202202130 ↗
13. 10.1093/nsr/nwy13510.1093/nsr/nwy135 ↗
14. 10.1126/sciadv.abn790510.1126/sciadv.abn7905 ↗
15. 10.1093/nsr/nwad16410.1093/nsr/nwad164 ↗
16. 10.1038/s41928-021-00589-710.1038/s41928-021-00589-7 ↗
17. 10.1038/s41566-021-00793-z10.1038/s41566-021-00793-z ↗
18. 10.1126/science.abq868410.1126/science.abq8684 ↗
19. 10.1088/2040-8986/ad065910.1088/2040-8986/ad0659 ↗
20. 10.2528/PIER2112140510.2528/PIER21121405 ↗
21. 10.2528/PIER2202030810.2528/PIER22020308 ↗
22. 10.1002/978111864670010.1002/9781118646700 ↗
23. 10.1002/adom.20220074810.1002/adom.202200748 ↗
24. 10.1002/advs.20190012810.1002/advs.201900128 ↗
25. 10.1038/s41467-023-40619-w10.1038/s41467-023-40619-w ↗
26. 10.1038/s41565-022-01284-010.1038/s41565-022-01284-0 ↗
27. 10.2528/PIER2309020110.2528/PIER23090201 ↗
28. 10.1002/advs.20220469910.1002/advs.202204699 ↗
29. 10.1038/s41566-022-01126-410.1038/s41566-022-01126-4 ↗
30. 10.1117/1.AP.6.1.01600110.1117/1.AP.6.1.016001 ↗
31. 10.1515/nanoph-2023-056910.1515/nanoph-2023-0569 ↗
32. 10.1002/lpor.20220027410.1002/lpor.202200274 ↗
33. 10.1038/s41565-021-00895-310.1038/s41565-021-00895-3 ↗
34. 10.1038/s41578-022-00513-110.1038/s41578-022-00513-1 ↗
35. Krizhevsky A. · Proc. Adv. Neural Inf. Process. Syst. · 2012 · 25 · 1097
36. K.Cho B.vanMerrienboer C.Gulcehre D.Bahdanau F.Bougares H.Schwenk Y.Bengio arXiv arXiv:1406.1078 2014 1724.
37. 10.1038/nature1696110.1038/nature16961 ↗
38. 10.2528/PIER2206140310.2528/PIER22061403 ↗
39. 10.2528/PIER2206060110.2528/PIER22060601 ↗
40. 10.2528/PIER2205160110.2528/PIER22051601 ↗
41. 10.1038/s41467-021-23087-y10.1038/s41467-021-23087-y ↗
42. 10.1038/538020a10.1038/538020a ↗
43. 10.1016/j.jcp.2018.10.04510.1016/j.jcp.2018.10.045 ↗
44. 10.1038/s44172-023-00107-x10.1038/s44172-023-00107-x ↗
45. 10.1021/acs.nanolett.9b0185710.1021/acs.nanolett.9b01857 ↗
46. 10.1021/acsphotonics.3c0069410.1021/acsphotonics.3c00694 ↗
47. Staelin D. H. · 1994
48. 10.1109/TETCI.2017.276273910.1109/TETCI.2017.2762739 ↗
49. 10.1021/acsphotonics.7b0137710.1021/acsphotonics.7b01377 ↗
50. 10.1109/2.48589110.1109/2.485891 ↗
51. Vaswani A. · Proc. Adv. Neural Inf. Process. Syst. · 2017 · 30 · 5998
52. J.Nilsson T.Akenine‐Möller arXiv 2020 2006.13846.