Machine learning and evolutionary algorithm studies of graphene metamaterials for optimized plasmon-induced transparency

Abstract

Machine learning and optimization algorithms have been widely applied in the design and optimization for photonics devices. We briefly review recent progress of this field of research and show data-driven applications, including spectrum prediction, inverse design and performance optimization, for novel graphene metamaterials (GMs). The structure of the GMs is well-designed to achieve the wideband plasmon induced transparency (PIT) effect, which can be theoretically demonstrated by using the transfer matrix method. Some traditional machine learning algorithms, including k nearest neighbour, decision tree, random forest and artificial neural networks, are utilized to equivalently substitute the numerical simulation in the forward spectrum prediction and complete the inverse design for the GMs. The calculated results demonstrate that all algorithms are effective and the random forest has advantages in terms of accuracy and training speed. Moreover, evolutionary algorithms, including single-objective (genetic algorithm) and multi-objective optimization (NSGA-II), are used to achieve the steep transmission characteristics of PIT effect by synthetically taking many different performance metrics into consideration. The maximum difference between the transmission peaks and dips in the optimized transmission spectrum reaches 0.97. In comparison to previous works, we provide a guidance for intelligent design of photonics devices based on machine learning and evolutionary algorithms and a reference for the selection of machine learning algorithms for simple inverse design problems.

Living Things Are Not (20th Century) Machines: Updating Mechanism Metaphors in Light of the Modern Science of Machine Behavior

“…Somewhat surprisingly, economic theory provided one of the first intuition pumps for considering the nonhuman generators of machines: machines arise from the literal hands of human engineers but also the "invisible hand" of a free market; the latter set of pressures in effect "select, " without human design or forethought, which technologies proliferate (Beinhocker, 2020). More recently, evolutionary algorithms, a type of machine learning algorithm, have demonstrated that, among other things, jet engines (Yu et al, 2019), metamaterials (Zhang et al, 2020), consumer products (Zhou et al, 2020), robots (Brodbeck et al, 2018;Shah et al, 2020), and synthetic organisms (Kriegman et al, 2020) can be evolved rather than designed: an evolutionary algorithm generates a population of random artifacts, scores them against human-formulated desiderata, and replaces low-scoring individuals with randomly modified copies of the survivors. Indeed, the "middle man" has even been removed in some evolutionary algorithms by searching for novelty rather than selecting for a desired behavior (Lehman and Stanley, 2011).…”

“…All biological and artificial materials and machines strike careful but different balances between many competing performance requirements. By drawing on advances in chemistry, materials science, and synthetic biology, a wider range of material, chemical and biotic building blocks are emerging, such as metamaterials and active matter (Silva et al, 2014;Bernheim-Groswasser et al, 2018;McGivern, 2019;De Nicola et al, 2020;Pishvar and Harne, 2020;Zhang et al, 2020), novel chemical compounds (Gromski et al, 2020), and computerdesigned organisms (Kriegman et al, 2020). These new building blocks may in turn allow artificial or natural evolutionary pressures to design hybrid systems that set new performance records for speed, dexterity, metabolic efficiency, or intelligence, while easing unsatisfying metabolic, biomechanical and adaptive tradeoffs.…”

Deep Learning the Electromagnetic Properties of Metamaterials—A Comprehensive Review

“…[ 123–125 ] In photonic crystals, DNNs have been used to optimize the Q‐factor in nanocavities, [ 99 ] waveguide properties in fibers, [ 126 ] compute the band structure in 1D [ 127 ] and 2D [ 128–130 ] PCs, and predict edge states in topological insulators. [ 107 ] Last, several groups have utilized DNNs in the optimization of nanophotonic devices including plasmonic [ 131,132 ] and dielectric [ 102,133–137 ] waveguides, nanoantennas, [ 101,110 ] thermophotovoltaics, [ 138 ] power splitters, [ 133 ] biosensors, [ 139 ] smart windows, [ 140 ] and grating couplers. [ 141–143 ]…”

Efficient Optical Spatial First-Order Differentiator Based on Graphene-Based Metalines and Evolutionary Algorithms

“…In addition, the high tunability of graphene metalines can be combined with inverse design technology to design a more efficient photonics device [28]- [30]. In general, inverse design is converted to the optimization problem which can be solved by gradient based methods (such as adjoint variable method (AVM) [31]- [32]), gradient free methods (such as genetic algorithm (GA) [33]- [37]) and model based methods (such as machine learning [38]- [40]). As a representative algorithm of gradient based methods, AVM not only designs linear devices but also optimizes for the nonlinear devices in frequency domain, but it requires physical background to derive the gradient of objective function [31]- [32].…”

“…As a representative algorithm of gradient based methods, AVM not only designs linear devices but also optimizes for the nonlinear devices in frequency domain, but it requires physical background to derive the gradient of objective function [31]- [32]. Apart from the gradient-based methods, model based methods, such as artificial neural networks and random forest, are also used to inversely design the photonics devices [38]- [40]. However, in order to train the model whose inputs are physical parameters and outputs are electromagnetic responses, it requires a significant amount of time to generate the training instances and test instances [38].…”