☆ 4.7 Article

Fabrication-conscious neural network based inverse design of single-material variable-index multilayer films

NANOPHOTONICS (2023)

Related references

Note: Only part of the references are listed.

Article Multidisciplinary Sciences

Ti/TiO2/SiO2 multilayer thin films with enhanced spectral selectivity for optical narrow bandpass filters

Dongju Kim et al.

Summary: Thin film-based optical sensors have attracted increasing interest in biometrics and other developing technologies. This study focuses on using multilayered dielectric thin films to modulate the optical properties and improve the spectral selectivity of Thin Film Narrow Bandpass Filters (TFNBFs). By incorporating metallic thin films, narrower bandwidths with higher transmittance have been achieved. Computer simulations were used to determine the optimal number and thickness of layers in Ti/TiO2/SiO2 based thin films, which efficiently tailor the optical path transmitting specific wavelength bands. The addition of Ti metal layers significantly changed the cutoff frequency of transmittance at specific wavelengths. This research contributes to the future development of simulation-based design strategies for industrial applications such as optical biometrics.

SCIENTIFIC REPORTS (2022)

Add to Collection

Article Optics

TMM-Fast, a transfer matrix computation package for multilayer thin-film optimization: tutorial

Alexander Luce et al.

Summary: This article introduces a Python package for calculating optical reflection and transmission in multilayer thin film structures, which provides fast parallel computation for experimentation with new optimization techniques, generation of datasets for machine learning, and effective evolutionary optimization. Additionally, an OpenAI Gym environment is provided for training reinforcement learning agents on the problem of finding multilayer thin-film configurations.

JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION (2022)

Add to Collection

Article Optics

Data-driven design of thin-film optical systems using deep active learning

Youngjoon Hong et al.

Summary: A deep learning aided optimization algorithm is developed for the design of flat thin-film multilayer optical systems, achieving near-optimal solutions for the inverse design problem of creating anti-reflective gratings. The algorithm demonstrates superior performance and lower computational cost compared to traditional nonlinear optimization algorithms. Additionally, a regression model using deep active learning is developed to accurately predict the total reflectivity of optical systems, allowing for rapid evaluation and design.

OPTICS EXPRESS (2022)

Add to Collection

Article Optics

Statistical information enhanced robust design method of optical thin film

Jianglin Dai et al.

Summary: A novel merit function was proposed based on the spectral coefficient average error and standard deviation, and a multi-objective optimization strategy was used for optimization. The control effect of the new algorithm on the sensitivity of film parameters error was verified by comparing the results of different filters designed by the conventional algorithm and the new algorithm.

OPTICS EXPRESS (2022)

Add to Collection

Review Optics

Deep learning for the design of photonic structures

Wei Ma et al.

Summary: Innovative approaches and tools, particularly deep learning, are shaping the field of photonics by offering efficient means to design photonic structures and providing data-driven solutions complementary to traditional physics-based methods. The progress in deep-learning-based photonic design is promising, with various model architectures showing potential applications in specific photonic tasks.

NATURE PHOTONICS (2021)

Add to Collection

Article Nanoscience & Nanotechnology

Multiobjective and categorical global optimization of photonic structures based on ResNet generative neural networks

Jiaqi Jiang et al.

Summary: The study demonstrates that deep generative neural networks based on global optimization networks can efficiently perform the multiobjective and categorical global optimization of photonic devices, pushing the capabilities of inverse design algorithms for photonics.

NANOPHOTONICS (2021)

Add to Collection

Article Nanoscience & Nanotechnology

General Inverse Design of Layered Thin-Film Materials with Convolutional Neural Networks

Andrew Lininger et al.

Summary: The study demonstrates the application of convolutional neural networks in solving the inverse design problem for metamaterials made of stacks of thin films, showcasing the remarkable ability of neural networks to navigate large design spaces and establish relationships between structure and optical responses. Compared to traditional optimization methods, the relative efficiency of neural networks increases with the total layer number, highlighting the advantage of machine learning approaches in complex, multi-layered systems.

ACS PHOTONICS (2021)

Add to Collection

Article Multidisciplinary Sciences

Room-temperature single-photon emitters in silicon nitride

Alexander Senichev et al.

Summary: Single-photon emitters in silicon nitride (SiN) films grown on silicon dioxide substrates have been observed at room temperature, exhibiting high brightness, stability, linear polarization, and purity. These emitters could originate from specific defect centers in SiN, showing potential for direct, scalable, and low-loss integration with established photonic on-chip platforms.

SCIENCE ADVANCES (2021)

Add to Collection

Article Physics, Applied

Parameterized reinforcement learning for optical system optimization

Heribert Wankerl et al.

Summary: In this study, a Q-learning variant is proposed to solve the inverse design optimization, outperforming human experts and current methods in optimizing multi-layer thin film structures. An exponentially transformed reward signal is introduced to simplify policy search and enable constrained optimization. The learned Q-values also provide information on the optical properties of multi-layer thin films, allowing for physical interpretation and what-if analysis.

JOURNAL OF PHYSICS D-APPLIED PHYSICS (2021)

Add to Collection

Article Nanoscience & Nanotechnology

Machine learning-assisted global optimization of photonic devices

Zhaxylyk A. Kudyshev et al.

Summary: This study introduces a new method for designing photonic metadevices using machine learning, which significantly improves the optimization efficiency of device configurations with complex topologies by combining adversarial autoencoder and metaheuristic optimization framework. The approach involves physics-driven compressed design space engineering to further optimize designs based on optical responses.

NANOPHOTONICS (2021)

Add to Collection

Article Nanoscience & Nanotechnology

Inverse Design of Photonic Crystals through Automatic Differentiation

Momchil Minkov et al.

ACS PHOTONICS (2020)

Add to Collection

Article Physics, Applied

Machine-learning-assisted metasurface design for high-efficiency thermal emitter optimization

Zhaxylyk A. Kudyshev et al.

APPLIED PHYSICS REVIEWS (2020)

Add to Collection

Article Multidisciplinary Sciences

Multilayer optical thin film design with deep Q learning

Anqing Jiang et al.

SCIENTIFIC REPORTS (2020)

Add to Collection

Article Materials Science, Multidisciplinary

Thin SiNC/SiOC Coatings with a Gradient of Refractive Index Deposited from Organosilicon Precursor

Hieronim Szymanowski et al.

COATINGS (2020)

Add to Collection

Article Nanoscience & Nanotechnology

Simulator-based training of generative neural networks for the inverse design of metasurfaces

Jiaqi Jiang et al.

NANOPHOTONICS (2020)

Add to Collection

Article Nanoscience & Nanotechnology

Design and fabrication of robust broadband extreme ultraviolet multilayers

Shang-qi Kuang et al.

AIP ADVANCES (2019)

Add to Collection

Article Chemistry, Multidisciplinary

Global Optimization of Dielectric Metasurfaces Using a Physics-Driven Neural Network

Jiaqi Jiang et al.

NANO LETTERS (2019)

Add to Collection

Article Optics

Robust inverse-design of scattering spectrum in core-shell structure using modified denoising autoencoder neural network

Baiqiang Hu et al.

OPTICS EXPRESS (2019)

Add to Collection

Article Nanoscience & Nanotechnology

Forward-Mode Differentiation of Maxwell's Equations

Tyler W. Hughes et al.

ACS PHOTONICS (2019)

Add to Collection

Article Optics

Robust design of broadband EUV multilayer using multi-objective evolutionary algorithm

Shang-qi Kuang et al.

OPTICS COMMUNICATIONS (2018)

Add to Collection

Article Nanoscience & Nanotechnology

Training Deep Neural Networks for the Inverse Design of Nanophotonic Structures

Dianjing Liu et al.

ACS PHOTONICS (2018)

Add to Collection

Article Materials Science, Multidisciplinary

Investigation of solar cell performance using multilayer thin film structure (SiO2/Si3N4) and grating

R. S. Dubey et al.

RESULTS IN PHYSICS (2017)

Add to Collection

Article Computer Science, Artificial Intelligence

Mean Absolute Percentage Error for regression models

Arnaud de Myttenaere et al.

NEUROCOMPUTING (2016)

Add to Collection

Article Optics

Antireflection coatings for multilayer-type photonic crystals

Yasuo Ohtera et al.

OPTICS EXPRESS (2010)

Add to Collection

Article Optics

Optical coating design approaches based on the needle optimization technique

A. V. Tikhonravov et al.

APPLIED OPTICS (2007)

Add to Collection

Article Materials Science, Coatings & Films

Graded refractive index layer systems for antireflective coatings and rugate filters deposited by reactive pulse magnetron sputtering

H Bartzsch et al.

SURFACE & COATINGS TECHNOLOGY (2004)

Add to Collection

Article Optics

Single-material inhomogeneous optical filters based on microstructural gradients in plasma-deposited silicon nitride

R Vernhes et al.

APPLIED OPTICS (2004)

Add to Collection

© Peeref 2019-2024. All rights reserved.