Inverse design of an on-chip optical response predictor enabled by a deep neural network

Kim, Junhyeong; Neşeli, Berkay; Kim, Jae-Yong; Yoon, Jinhyeong; Yoon, Hyeonho; Park, Hyo‐Hoon; Kurt, Hamza

doi:10.1364/oe.480644

Cited by 15 publications

(4 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[ 150 ] Figure 20c [ 151 ] and d [ 152 ] present another example of developing a grating coupler with deep learning, including forward modeling and inverse design. Lately, the FCN has extended to a wider range of communication‐relevant photonic platforms, including Bragg grating, [ 153 ] directional coupler, [ 154 ] nanophotonic waveguide, [ 155 ] and power splitter, [ 156 ] as sketched in Figure 20e–h.…”

Section: Deep Learning In the Acceleration Of Silicon Photonics Researchmentioning

confidence: 99%

Section: Deep Learning In the Acceleration Of Silicon Photonics Researchmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

The Intelligent Design of Silicon Photonic Devices

Li,

Zhou,

Qiu

et al. 2024

Advanced Optical Materials

View full text Add to dashboard Cite

Photonic devices based on silicon waveguides are essential to versatile high‐performance and low‐cost photonic integrated systems. Extremely complex silicon photonic devices with hundreds or even thousands of degrees of freedom (DOF) are successfully designed and manufactured based on recent advances in data science and nanofabrication technology. At this level, conventional forward‐reasoning may no longer be suitable for designing high‐performance silicon photonic devices with novel functionalities since the light‐matter interaction is complex and non‐intuitive. Therefore, the timely development of sub‐wavelength silicon photonic devices that can precisely mold the flow of light is a critical and urgent issue requiring joint engineering and scientific efforts. In this paper, an inverse design strategy based on heuristic and gradient descendant algorithms, enabling the realization of large‐scale integrated devices is first introduced. Subsequently, the burgeoning deep learning technology, which offers a promising direction for the automation design of silicon photonics with a data‐driven approach, is discussed. Finally, the obstacles and prospects in this emerging research direction are revealed. Detail discussions from multiple perspectives are provided. This review aims to provide general guidance and a comprehensive reference for scientists developing photonic integrated systems.

show abstract

Section: Deep Learning In the Acceleration Of Silicon Photonics Researchmentioning

confidence: 99%

Section: Deep Learning In the Acceleration Of Silicon Photonics Researchmentioning

confidence: 99%

See 1 more Smart Citation

The Intelligent Design of Silicon Photonic Devices

Li,

Zhou,

Qiu

et al. 2024

Advanced Optical Materials

View full text Add to dashboard Cite

show abstract

“…After training FNN (the spectrum prediction network), the hyperparameters of the INN (the design parameters prediction network) are also similarly optimized as hyperparameters are shared in Table 2. To ensure spectral response prediction and design parameters prediction of the entire network are close to the indicated values in the training dataset, Kim et al [17] introduced weighted loss function by adding design parameters loss to total loss calculation of network recently. The equation below is the loss function used to train our TNN.…”

Section: Table 2 Hyperparameters Of Deep Neural Networkmentioning

confidence: 99%

Artificial intelligence technique aided rapid design of THz stereo-metamaterial-polarizer and sensing application

Güngördü

Sanders

Rahman

et al. 2023

Next-Generation Spectroscopic Technologies XV

View full text Add to dashboard Cite

Polarization conversion devices which can manipulate the polarization status of electromagnetic waves are essential to various areas of photonic applications such as communication, imaging, and remote sensing. Due to its wide bandwidth and high penetration through dielectric materials, THz wave, range from 0.1 to 10 THz is increasingly popular for noninvasive screening, high-resolution imaging, and more precise data collection. Metamaterials (MM) are artificial materials fabricated from repeated arrays of subwavelength-sized meta-atoms, and MM-based THz polarization conversion devices can be thin, extremely compact, easy to integrate, and even flexible, unlike conventional polarization devices. In light of that information, we utilized a stereo-metamaterial (SMM) structure for new-generation THz polarizers converting the polarization from linearly to circularly and elliptically polarized wave at the THz frequency range in reflection mode. In this work, we present the processes and results of the inverse design of SMM using the artificial neural network (ANN), trained by various parameters, including polarization status and ellipticity angle, to achieve highly efficient device performance. Training and testing our ANN with the created datasets by simulation for the inverse design of the device, design parameters were obtained by giving an artificial EM response or ellipticity angle spectrum or vice versa more efficiently and rapidly. With the device fabricated based on the ANN-powered design, we demonstrated effective sensing of different polarization statuses using THz polarimetry spectroscopy.

show abstract

Semi‐Supervised Learning Leveraging Denoising Diffusion Probabilistic Models for the Characterization of Nanophotonic Devices

Kim,

Neseli,

Yoon

et al. 2024

Laser & Photonics Reviews

View full text Add to dashboard Cite

Albeit the recent successful demonstrations of nanophotonic device designs leveraging data‐driven supervised learning methods, several challenges still remain. One of the primary constraints of these methods is their computational cost because they require a massive amount of highly time‐consuming full‐wave electromagnetic simulations. In this study, semi‐supervised learning methods are implemented to avoid this issue. Photonic crystal waveguide devices, which offer interesting light‐matter interactions such as the slow‐light effect and the filtering effect, are employed to validate the methodologies. To utilize unlabeled data within this framework, a novel deep generative model, the denoising diffusion probabilistic model is introduced. Next, a pseudo‐labeling process is introduced to assign synthetic labels to the unlabeled data. The performance of a semi‐supervised learning method is compared with a supervised learning scheme, demonstrating that the performance of the neural network can be enhanced without additional numerical calculations. This method holds extensive potential to elevate the efficiency of designing and characterizing future nanophotonic devices.

show abstract

Inverse design of an on-chip optical response predictor enabled by a deep neural network

Cited by 15 publications

References 36 publications

The Intelligent Design of Silicon Photonic Devices

The Intelligent Design of Silicon Photonic Devices

Artificial intelligence technique aided rapid design of THz stereo-metamaterial-polarizer and sensing application

Semi‐Supervised Learning Leveraging Denoising Diffusion Probabilistic Models for the Characterization of Nanophotonic Devices

Contact Info

Product

Resources

About