Deep neural networks for the evaluation and design of photonic devices

Jiang, Jiaqi; Chen, Mingkun; Fan, Jonathan A.

doi:10.1038/s41578-020-00260-1

Cited by 381 publications

(246 citation statements)

References 129 publications

Supporting

Mentioning

244

Contrasting

Order By: Relevance

“…Recently, it has also shown the potential for other applications such as sensing [4] and light detection and ranging (LiDAR) [5]. For miscellaneous silicon photonic devices, their versatile functions mainly come from design of device geometry [6]. Intuitively, the design of single component with specific function is fulfilled by physics-based methods like analytical models, prior practical experiences and scientific intuitions [7].…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, the performance evaluation of lightmaterial interaction with complicated geometries has to rely on numerical electromagnetic (EM) simulations [8], as the computation for irregular geometry is often non-intuitive. To make the design process more efficient, inverse design approaches have been introduced, which are assisted by various iterative optimization methods and deep neural networks (DNNs) [6][7][8][9][10][11][12].…”

Section: Introductionmentioning

confidence: 99%

“…This is extremely laborious when many similar devices are needed with slightly difference target response, such as power splitters with different splitting ratios or grating couplers (GCs) working at different wavelength bands [16]. Fortunately, DNN, an unprecedented tool capable of building a bridge between data features and data labels, has also found its way to assist the design of silicon photonic devices [6]. Based on the mapping direction between geometric parameters and optical response, DNNs are trained as forward models and inverse models.…”

Section: Introductionmentioning

confidence: 99%

“…For example, Molesky et al focused on the adjoint method and its applications [7] and Jensen et al concentrated on topology optimization of nanophotonics [15]. More recently, Jiang et al and So et al presented very detailed reviews about DNNs-assisted metasurface and nanophotonics design [6,8]. Different from these previous reviews, we review design methodologies for silicon photonics from the perspective of demands.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Inverse Design for Silicon Photonics: From Iterative Optimization Algorithms to Deep Neural Networks

et al. 2021

View full text Add to dashboard Cite

Silicon photonics is a low-cost and versatile platform for various applications. For design of silicon photonic devices, the light-material interaction within its complex subwavelength geometry is difficult to investigate analytically and therefore numerical simulations are majorly adopted. To make the design process more time-efficient and to improve the device performance to its physical limits, various methods have been proposed over the past few years to manipulate the geometries of silicon platform for specific applications. In this review paper, we summarize the design methodologies for silicon photonics including iterative optimization algorithms and deep neural networks. In case of iterative optimization methods, we discuss them in different scenarios in the sequence of increased degrees of freedom: empirical structure, QR-code like structure and irregular structure. We also review inverse design approaches assisted by deep neural networks, which generate multiple devices with similar structure much faster than iterative optimization methods and are thus suitable in situations where piles of optical components are needed. Finally, the applications of inverse design methodology in optical neural networks are also discussed. This review intends to provide the readers with the suggestion for the most suitable design methodology for a specific scenario.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Inverse Design for Silicon Photonics: From Iterative Optimization Algorithms to Deep Neural Networks

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The inverse design of nanophotonic structures, meaning the design of an optimal metamaterial given a desired spectral response, would be of great commercial value, accelerating material development and the discovery of new structures. Due to the high dimensionality of the phenomena responsible for structural color, inverse design remains a challenge 12 . Conventional design workflows employ numerical simulations based on Maxwell's equations to accurately solve a) mehdi.keshavarz-hedayati@durham.ac.uk the EM response, given a metamaterial structure (forward mapping) 13 .…”

mentioning

confidence: 99%