End-to-End Diverse Metasurface Design and Evaluation Using an Invertible Neural Network
Nanomaterials · 15.09.2023 · 13 (18) · 2561
DOI: 10.3390/nano13182561 ↗Kurz gefasst
Fachartikel in der Zeitschrift Nanomaterials, online erschienen am 15.09.2023. Verfasst von Yunxiang Wang und 7 weiteren Autor:innen. Beteiligte Einrichtung: Ming Hsieh Department of Electrical Engineering, University of Southern California, Los Angeles, CA 90089, USA. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Innovusion Inc, Innovusion Inc. via the “Optical metalenses for LiDAR” program.
Abstract (Originalsprache)
Employing deep learning models to design high-performance metasurfaces has garnered significant attention due to its potential benefits in terms of accuracy and efficiency. A deep learning-based metasurface design framework typically comprises a forward prediction path for predicting optical responses and a backward retrieval path for generating geometrical configurations. In the forward design path, a specific geometrical configuration corresponds to a unique optical response. However, in the inverse design path, a single performance metric can correspond to multiple potential designs. This one-to-many mapping poses a significant challenge for deep learning models and can potentially impede their performance. Although representing the inverse path as a probabilistic distribution is a widely adopted method for tackling this problem, accurately capturing the posterior distribution to encompass all potential solutions remains an ongoing challenge. Furthermore, in most pioneering works, the forward and backward paths are captured using separate models. However, the knowledge acquired from the forward path does not contribute to the training of the backward model. This separation of models adds complexity to the system and can hinder the overall efficiency and effectiveness of the design framework. Here, we utilized an invertible neural network (INN) to simultaneously model both the forward and inverse process. Unlike other frameworks, INN focuses on the forward process and implicitly captures a probabilistic model for the inverse process. Given a specific optical response, the INN enables the recovery of the complete posterior over the parameter space. This capability allows for the generation of novel designs that are not present in the training data. Through the integration of the INN with the angular spectrum method, we have developed an efficient and automated end-to-end metasurface design and evaluation framework. This novel approach eliminates the need for human intervention and significantly speeds up the design process. Utilizing this advanced framework, we have effectively designed high-efficiency metalenses and dual-polarization metasurface holograms. This approach extends beyond dielectric metasurface design, serving as a general method for modeling optical inverse design problems in diverse optical fields.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Ming Hsieh Department of Electrical Engineering, University of Southern California, Los Angeles, CA 90089, USA.
- The High School Affiliated to Renmin University of China, CUIWEI Campus, Beijing 100086, China.
ORCID-Profile: Yunxiang Wang, Tse-Hsien Ou, Wei Wu
Förderung
- Innovusion Inc
- Innovusion Inc. via the “Optical metalenses for LiDAR” program
Zitieren
APA
Wang, Y., Yang, Z., Hu, P., Hossain, S., Liu, Z., Ou, T. H., Ye, J., & Wu, W. (2023). End-to-End Diverse Metasurface Design and Evaluation Using an Invertible Neural Network. Nanomaterials, 13(18), 2561. https://doi.org/10.3390/nano13182561
Vancouver
Wang Y, Yang Z, Hu P, Hossain S, Liu Z, Ou TH, et al. End-to-End Diverse Metasurface Design and Evaluation Using an Invertible Neural Network. Nanomaterials. 2023;13(18):2561. doi:10.3390/nano13182561
BibTeX
@article{Wang2023,
author = {Wang, Yunxiang and Yang, Ziyuan and Hu, Pan and Hossain, Sushmit and Liu, Zerui and Ou, Tse-Hsien and Ye, Jiacheng and Wu, Wei},
title = {End-to-End Diverse Metasurface Design and Evaluation Using an Invertible Neural Network},
journal = {Nanomaterials},
year = {2023},
volume = {13},
number = {18},
pages = {2561},
doi = {10.3390/nano13182561}
}
EUVIMED