Metaphotonics gets smarter | Eurek alert!





Figure 1

image: Reverse and direct design based on DL techniques.
to see Continued

Credit: OAE

A new publication from Advances in optoelectronics; DOI 10.29026/oea.2022.210147 presents the evaluation of artificial intelligence-induced metaphotonics and presents a summary of machine learning concepts with some specific examples developed and demonstrated for metasystems and metasurfaces.

Advances in the field of artificial intelligence have resulted in the incorporation of these technologies into the scientific research process and into the field of photonics. Methods such as machine learning and deep learning have become popular design tools for the development of photonic devices. In this case, design involves predicting a physical response of a given structure (direct design) as well as the inverse process of finding the parameters of a structure needed to provide a desired response (inverse design). While design procedures remain arguably the most widespread implementation of machine learning in photonics, new applications are beginning to emerge, leading to the evolution of a new research area of smart photonics.

The authors of this article review recent advances in the field of intelligent metaphotonics. This subfield of photonics is dedicated to intelligent systems and devices driven by optically induced electrical and magnetic resonances obtained through the use of sub-wavelength structured nanoparticles (commonly called meta-atoms). The review covers machine learning for the design of such devices as well as other uses of machine learning, including classification tools, control systems, and feedback mechanisms. Particular attention is devoted to potential practical applications, including solar cells, biosensors or imagers. Avoiding diving into a detailed description of the methods, the authors connect various aspects of machine learning and metaphotonics by creating a broad and coherent picture of their interaction, highlighting the peculiarities of intelligent systems. New concepts such as self-adaptive systems or smart biosensors are introduced and discussed in the context of the future development of metaphotonics. Under self-adaptation, the authors imply the ability of devices to automatically adjust their responses with changing environmental conditions. One of the examples provided is a cloak that adjusts to changes in frequency and angle of incidence of the electromagnetic field, which allows functionality to be preserved in a wide range of conditions. The concept of smart biosensors involves the use of machine learning as a tool for classifying samples. The review also highlights the importance of metaphotonics for artificial intelligence by demonstrating several examples of metasurfaces used as platforms for the all-optical realization of machine learning algorithms.

For the general public, the review may be of interest as an introduction to the field of intelligent metaphotonics as well as an overview of its state of the art. No specific computer knowledge is required as the authors provide a brief description of the ideas behind ML and avoid technical details. The article provides descriptive explanations of emerging concepts to familiarize readers with the new direction of photonics development and provide references for further in-depth study.

The authors of this article review the field of intelligent metaphotonics – scientific field at the junction of artificial intelligence (AI) and metaphotonics.

AI is quickly becoming part of work and everyday life all over the world. Recent advances in this field have become possible in large part due to the rise of a concept such as machine learning (ML) representing a group of data-driven algorithms for AI learning capability. . It can be said that ML allows the computer to learn from experience, which means that with the increased amount of input data, a program can better solve its predefined tasks. In the same way, repetitions and training improve human performance, except that the learning mechanisms and abilities are quite different.

Science does not stay away from modern trends and adopts advanced computational methods for solving problems and developing new concepts. This article focuses on the application of ML in metaphotonics – a burgeoning field of subwave photonics inspired by metamaterial physics. Initially presented as a tool for the forward and reverse design of photonic systems, ML has already evolved into something more than just a design method. The review provides examples of how ML has been integrated as a support tool for photonic sensors or the feedback and control mechanism of self-adaptive systems. Without a doubt, ML is rapidly becoming a powerful tool for research in the field of metaphotonics. The merger of these two fields has led to the development of a new area of ​​research commonly referred to as intelligent metaphotonics which involves metaphotonic systems designed or improved with ML or AI in general.

The article provides a brief introduction to AI concepts, throughout the article ML is treated as a “black box” providing a desired result without specific details on how exactly it was achieved. Such an approach allows authors to focus on specific metaphotonic systems and their ML-enabled applications rather than specific methods and algorithms. Also, in this way, the document may be of interest to readers who are not familiar with computers. After the introduction, the article covers the ML-aided design of nanoantennas which are building blocks of metaphotonic structures. Next, the focus is on transformative metasurfaces and improving their properties with ML. Particular attention is devoted to potential real-world applications, such as structural colors, LIDAR or near-eye displays. The applications of metasurfaces as biosensing platforms are presented in a separate section, since in this field ML can be used not only as a design method, but also as a tool for classifying samples. This idea is demonstrated with examples of sensors for the classification of SARS-CoV-2 and to monitor the dynamics of biomolecules. Another class of applications involves self-adaptive devices – metaphotonic systems that can automatically adapt to changes in the environment and adjust their responses. Examples provided include self-adaptive microwave capes and imagers as well as metaphotonic systems used as computational platforms for ML. Finally, the authors provide a perspective covering emerging and potential trends in the field of smart metaphotonics assuming not only the influence of ML on photonics, but also the inverse importance of photonics for AI technologies.

Article reference: Krasikov S, Tranter A, Bogdanov A, Kivshar Y. Smart metaphotonics enhanced by machine learning. Opto-Electron Adv 5, 210147 (2022). 10.29026/oea.2022.210147

Key words: metaphotonics; machine learning; artificial intelligence

# # # # # #

The research was led by Professor Yuri Kivshar, who is a world leader in photonics and metamaterials, and one of the founders of the field of all-dielectric resonant metaphotonics governed by the physics of Mie resonances in dielectric nanoparticles at high refractive index.

# # # # # #

Advances in optoelectronics (OEA) is a high-impact, open-access, peer-reviewed SCI monthly journal with an impact factor of 9.682 (Journals Citation Reports for IF 2020). Since its launch in March 2018, OEA has been indexed in SCI, EI, DOAJ, Scopus, CA and ICI databases over time and has expanded its editorial board to 36 members from 17 countries and regions (average h-index of 49).

The journal is published by the Institute of Optics and Electronics of the Chinese Academy of Sciences, aiming to provide a platform for researchers, scholars, professionals, practitioners and students to impart and share knowledge under the form of high quality empirical and theoretical research papers covering the topics of optics, photonics and optoelectronics.

# # # # # #

More information: http://www.oejournal.org/oea

Editorial Board: http://www.oejournal.org/oea/editorialboard/list

All issues are available in the online archive (http://www.oejournal.org/oea/archive).

Submissions to AEO can be done using ScholarOne (https://mc03.manuscriptcentral.com/oea).

ISSN: 2096-4579

NC: 51-1781/TN

Contact us: oea@ioe.ac.cn

Twitter: @OptoElectronAdv (https://twitter.com/OptoElectronAdv?lang=en)

WeChat: OE_Journal

# # # # # #


Warning: AAAS and EurekAlert! are not responsible for the accuracy of press releases posted on EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.





Leave a Comment

x