Reliable, efficient, and scalable photonic inverse design empowered by physics‐inspired deep learning
Open Access
- 27 January 2025
- journal article
- research article
- Published by Wiley in Nanophotonics
- Vol. 14 (16) , 2799-2810
Abstract
On‐chip computing metasystems composed of multilayer metamaterials have the potential to become the next‐generation computing hardware endowed with light‐speed processing ability and low power consumption but are hindered by current design paradigms. To date, neither numerical nor analytical methods can balance efficiency and accuracy of the design process. To address the issue, a physics‐inspired deep learning architecture termed electromagnetic neural network (EMNN) is proposed to enable an efficient, reliable, and flexible paradigm of inverse design. EMNN consists of two parts: EMNN Netlet serves as a local electromagnetic field solver; Huygens–Fresnel Stitch is used for concatenating local predictions. It can make direct, rapid, and accurate predictions of full‐wave field based on input fields of arbitrary variations and structures of nonfixed size. With the aid of EMNN, we design computing metasystems that can perform handwritten digit recognition and speech command recognition. EMNN increases the design speed by 17,000 times than that of the analytical model and reduces the modeling error by two orders of magnitude compared to the numerical model. By integrating deep learning techniques with fundamental physical principle, EMNN manifests great interpretability and generalization ability beyond conventional networks. Additionally, it innovates a design paradigm that guarantees both high efficiency and high fidelity. Furthermore, the flexible paradigm can be applicable to the unprecedentedly challenging design of large‐scale, high‐degree‐of‐freedom, and functionally complex devices embodied by on‐chip optical diffractive networks, so as to further promote the development of computing metasystems.Keywords
Funding Information
- National Natural Science Foundation of China (62088102)
- National Natural Science Foundation of China (62125106)
- China Postdoctoral Science Foundation (GZB20230372)
- China Association for Science and Technology (2023QNRC001)
This publication has 50 references indexed in Scilit:
- Fast Predicting the Complex Nonlinear Dynamics of Mode‐Locked Fiber Laser by a Recurrent Neural Network with Prior Information FeedingLaser & Photonics Reviews, 2023
- Real‐Time On‐Demand Design of Circuit‐Analog Plasmonic Stack Metamaterials by Divide‐and‐Conquer Deep LearningLaser & Photonics Reviews, 2022
- Physics‐Informed Neural Network for Nonlinear Dynamics in Fiber OpticsLaser & Photonics Reviews, 2022
- Compounding Meta‐Atoms into Metamolecules with Hybrid Artificial Intelligence TechniquesAdvanced Materials, 2019
- A Bidirectional Deep Neural Network for Accurate Silicon Color DesignAdvanced Materials, 2019
- Probabilistic Representation and Inverse Design of Metamaterials Based on a Deep Generative Model with Semi‐Supervised Learning StrategyAdvanced Materials, 2019
- Adjoint Method and Inverse Design for Nonlinear Nanophotonic DevicesACS Photonics, 2018
- Generative Model for the Inverse Design of MetasurfacesNano Letters, 2018
- Training Deep Neural Networks for the Inverse Design of Nanophotonic StructuresACS Photonics, 2018
- Vector Vortex Beam Generation with a Single Plasmonic MetasurfaceACS Photonics, 2016