Review for artificial intelligence-based micro-nanophotonic devices

Abstract

Nanophotonic is an emerging frontier that explores the interaction between light and matter at the nanoscale. The design of ultra-compact and high-performance photonic devices remains a central challenge in this field. Conventional forward design approaches, which often rely on empirical knowledge, are computationally demanding and offer limited flexibility. The advent of deep learning has enabled inverse design strategies that start from desired optical responses. This review first outlines the background of traditional photonic device design and introduces machine learning techniques, particularly deep learning, for modeling nanostructures. It then examines the application of advanced optimization methods, including topology optimization and genetic algorithms, in the design of devices such as meta-lens, meta-gratings, and planar beam splitters. These examples highlight the integration of optoelectronics with artificial intelligence. The interplay between light manipulation and intelligent algorithms is accelerating advancements in optical design, imaging, communications, and data analysis, opening new avenues for interdisciplinary innovation.

Graphical abstract

Keywords

artificial intelligence / deep learning / intelligent algorithms / optical neural networks / nanophotonic

1 Introduction

Owing to the unique electromagnetic properties of nanoscale photonic structures, light–matter interactions can be precisely engineered at the subwavelength scale [1−17]. Representative examples include localized surface plasmon resonances that confine light to deep-subwavelength volumes [18−20], the emergence of optical magnetic resonances in intrinsically non-magnetic materials [20−24], the tailored manipulation of optical near-fields through subwavelength architectures [25−27], and the occurrence of various nonlinear optical phenomena [28, 29]. These nanoscale optical effects can have a wide range of applications, such as invisibility cloaks [30−32], analog computing [33−36], wireless communications [37−44], integrated quantum optics [45−47], metamaterials [48−54], plasmons [55−60], photonic crystals [45, 61−63], and metasurfaces that can manipulate the amplitude, phase, and polarization of electromagnetic waves [16, 64−82] can provide control over light propagation. Reconfigurable metasurfaces represent a groundbreaking class of structures capable of dynamically manipulating electromagnetic waves through tuning mechanisms that utilize various approaches [83−94] or different materials [95−108] depending on the operating frequency.

Although optical devices such as near-field microscopes have advanced considerably since their invention [109–111], numerically modeling many problems associated with these devices remains challenging [112, 113]. Similar difficulties also arise in the design of nano-optical devices, where achieving specific functionalities often requires intensive simulations and parameter sweeps. These limitations highlight the need for more efficient strategies. Owing to its unique advantages, artificial intelligence [114−119] has found widespread applications across various fields [120−175], and it can likewise be leveraged to address the challenges encountered in nanophotonics. For instance, machine learning techniques such as genetic algorithms [141, 176] have been applied to optimize metasurface unit cells, while Bayesian optimization [177−180] has been employed to accelerate parameter tuning with reduced computational cost. In addition, deep learning methods [180−202] — including convolutional neural networks (CNNs), generative adversarial networks (GANs) [203, 204], and conditional generative models — have been successfully used for the inverse design of photonic structures, enabling direct prediction of geometry from target optical responses. These approaches not only reduce the design cycle from days to minutes but also unlock innovative device functionalities and high-performance metasurfaces.

Over the past decade, the convergence of nanophotonic and machine learning has emerged as a prominent and rapidly evolving research frontier. Early applications of machine learning in photonics include contributions to inverse quantum design [205−207], computational microscopy [208], sensing [209, 210], metrology [211], optical neural networks (ONNs) [212−214], and other areas. Among the various branches of machine learning, deep learning stands out as a transformative force driving the rapid advancement of artificial intelligence. Built upon neural network architectures, deep learning has become a dominant framework in modern machine learning research and applications. Its algorithms, inspired by the human learning process, have achieved remarkable success in numerous domains, including feature detection [135, 215−218], speech recognition [122, 130, 219−233], image processing [136, 137, 203, 234−274] and object classification [176, 275−288].

The inverse design process based on deep learning methods [134, 156, 289−300] primarily involves two components: the forward design stage of the neural network and the inverse design stage. At the core of deep learning technology lies the construction of artificial neural networks (ANNs). ANNs are inspired by the working principles of biological neurons, in which the fundamental unit of information processing is the artificial neuron or node. Each neuron receives multiple inputs, multiplies them by corresponding weights, adds a bias term, and then passes the result through a nonlinear activation function. This mechanism is analogous to the synaptic weights and soma processing in biological neurons, allowing neurons to flexibly determine whether to “activate” based on the input signal and transmit information to the next layer. Raw data are received by the input layer, while the output layer generates the final results. The hidden layers progressively extract abstract features through layer-by-layer nonlinear mapping, enabling deep learning to automatically learn complex functional representations. To train ANNs, electromagnetic simulations are typically used to generate large sets of training samples. Starting from a randomly initialized network, the output is computed through forward propagation. The weights are then iteratively optimized using the backpropagation algorithm [301−303], guided by a loss function (such as mean squared error (MSE), cross-entropy, or mean absolute error), until the model accurately approximates the target distribution of the training set.

In neural-network modeling of nanophotonic devices, input labels x denote structural parameters (geometry, materials, excitation) and output labels y denote device responses (spectra). Electromagnetics makes the forward mapping x→y deterministic and single-valued, so forward networks can learn this mapping accurately and act as fast surrogates for costly simulations. The inverse mapping y→x is inherently one-to-many and multimodal — a single response y can correspond to many different designs x — so inverse networks struggle to represent this ambiguity and often converge to averaged solutions that miss valid design diversity. Even cascade or tandem schemes that couple inverse and forward models to enforce physical consistency typically produce a single feasible design rather than enumerating alternative viable structures.

Compared with traditional models, the tandem neural network (TNN) [64, 304−306] integrates forward and inverse networks into a unified framework to efficiently address the one-to-many ambiguity problem in inverse design. In this approach, a forward model is first trained to predict the device’s physical response y from its physical parameters x, ensuring that the predictions remain consistent with physical laws. Building upon this, an inverse model is then trained, which takes the target response y as input and attempts to generate the corresponding parameters x. The generated x is subsequently fed back into the forward model for verification, thereby constraining the inverse output to ensure its validity and physical consistency. This tandem architecture not only enhances the stability of inverse design but also yields multiple physically valid solutions, making it an effective strategy for resolving the multi-solution ambiguity in complex photonic device design. In recent years, supported by deep learning and open-source frameworks such as TensorFlow [307, 308], the application of neural networks in photonics has expanded rapidly and achieved notable practical improvements.

Overall, this review systematically summarizes recent progress in applying artificial intelligence techniques to nanophotonic devices. As illustrated in Fig. 1, this review primarily centers on photonic devices, with particular emphasis on meta-lens, meta-grating, and beam splitters. From an algorithmic perspective, it addresses not only deep learning methods, including neural networks, but also a variety of optimization strategies, such as topology optimization (TO), genetic algorithms, and particle swarm optimization. Furthermore, Table 1 provides a comparative overview of commonly used machine learning algorithms, highlighting their respective advantages and limitations.

2 Nanophotonic devices

Photonics is the discipline that studies how light is generated, manipulated, and detected, and it plays a pivotal role in advancing modern technology. In 1960, Maiman demonstrated the first working laser [309, 310], marking the beginning of a new era in photonics. During the 1970s, the advent of low-loss optical fibers [311, 312] and the rise of fiber-optic communications [313, 314] enabled information to be transmitted at the speed of light; in particular, the development of sub-20 dB/km optical fibers by Corning Laboratory in 1970 became a landmark achievement [315−317]. In the 1980s, breakthroughs in nonlinear optics — such as frequency conversion technologies [318−324] — expanded the application of lasers to a wider range of fields. The 1990s witnessed the emergence of photonic crystals, which further accelerated the development of novel optical devices.

In the early 21st century, super-resolution imaging technologies overcame the diffraction limit, opening new avenues for optical microscopy and imaging. In the 2010s, the development of photonic integrated circuits (PICs) [325−329] enabled the integration of multiple optical components onto a single chip, thereby improving both speed and energy efficiency [330]. Entering the 2020s, the convergence of artificial intelligence and photonics has further accelerated the field, with AI being widely applied to optical design optimization, complex data processing, and intelligent sensing systems [177, 237, 331−343].

2.1 Meta-lens

The utilization of meta-lens is on the rise [344, 345], owing to their capacity to regulate phase distributions. Their functional capabilities have now surpassed those of conventional lenses. In Ref. [346], as shown in Fig. 2(a), Fan et al. proposed an improved genetic algorithm (GA) for synthesizing one-dimensional geometric phase-controlled dielectric metasurfaces by optimizing the target phase pattern, in contrast to earlier approaches that relied on amplitude control. The proposed GA workflow consists of initialization, random mutation, evolutionary screening, and replication. In this method, random mutations are introduced by altering the phase of individual unit cells, while a multi-step hierarchical mutation strategy is employed to overcome local optima, thereby enhancing the GA optimization process. The effectiveness of this approach is validated by generating a longitudinal optical pattern — specifically, a non-diffractive light sheet — using a one-dimensional geometric phase dielectric metasurface with a non-analytical, counterintuitive phase distribution. This strategy demonstrates the ability of the optimized GA to address complex metasurface design problems and to realize customized light-sheet generation. Reference [347], as shown in Fig. 2(b), presented an inverse-designed meta-optical element with a substantial numerical aperture of ~0.44 and an extended depth of field (EDOF), characterized by a lens-like point spread function (PSF). This EDOF device maintains the same focusing efficiency as a hyperbolic meta-lens throughout its entire focal range. By leveraging the extended focal range in combination with computational post-processing, the authors demonstrated achromatic broadband imaging across the entire visible spectrum using a compact 1 mm meta-optical system. Ref. [348], as shown in Fig. 2(c), also adopts an inverse design approach, demonstrating that axisymmetric structures based on full-wave Maxwell equations can address emerging challenges, such as large wavelength separations and the integration of active materials. This method enables the creation of fundamentally novel solutions distinct from conventional lenses. These solutions include reconfigurable optics that shift multi-frequency nodal points in response to variations in the refractive index, as well as multispectral lenses with significant spectral separation (λ = 1 μm and 10 μm). The inverse-designed metasurface achieved near-diffraction-limited focusing at the 1550 nm telecommunications wavelength, with a numerical aperture of 0.4. Due to axisymmetry, the full-wave Maxwell solver can be applied to structures with diameters spanning hundreds to thousands of wavelengths before domain decomposition becomes necessary. Moreover, multilayer TO with approximately 10⁵ degrees of freedom enables the solution of highly complex design problems, even within the constraints of axisymmetric geometries.

Another algorithm well-suited for meta-lens design is TO. As shown in Fig. 3(a), Lin et al. [349] proposed a fully Maxwell-based topological optimization of a single multilayer meta-lens with a thickness of approximately 10 wavelengths. This meta-lens achieves simultaneous focusing across 60 angular ranges and a 23% spectral bandwidth without chromatic or angular aberrations, thereby demonstrating “flat-field achromatism”. The system exhibits diffraction-limited performance and maintains an absolute focusing efficiency exceeding 50% across the range of incident angles and operating frequencies. Similarly, as shown in Fig. 3(b), Christiansen et al. [350] reported a genetic algorithm enhanced by topological optimization. Their study presents a concise code to illustrate the use of TO as an inverse design tool in photonics, exemplified through the design of two-dimensional dielectric meta-lens and metallic reflectors. The results further show that gradient-based optimization methods outperform genetic algorithm-based approaches for photonic inverse design, achieving a figure of merit (FOM) approximately 18% higher than that obtained with the genetic algorithm.

TO has also been extended to the design of large-area meta-lens. As shown in Fig. 3(c), Phan et al. [65] proposed a computationally efficient strategy in which individually optimized segments of a metasurface are stitched together, reducing the optimization complexity from a high-order polynomial to a linear scale. Using this approach, they designed and fabricated a large-area, high-numerical aperture silicon meta-lens that achieved focusing efficiencies exceeding 0.8. Similarly, as shown in Fig. 3(d), Lin et al. [351] proposed an overlapped-domain strategy for designing wide-area metasurfaces, in which each simulation unit includes a unit cell structure, overlapping regions from adjacent cells, and a perfectly matched layer (PML) absorber. Within this framework, they applied a topology modeling approach to design a large-area (200λ), high-numerical aperture (0.71) polychromatic, omnidirectional achromatic lens with a focusing efficiency of about 50%. The design incorporated a large number of degrees of freedom and was optimized to maximize the minimum focal intensity at ten evenly spaced wavelengths spanning the visible spectrum (480–700 nm).

Meta-lens designed using deep learning methodologies have attracted growing attention due to their potential for efficient and high-performance device optimization. As shown in Fig. 3(e), Pestourie et al. [68] proposed an active learning framework tailored for composite materials, which substantially accelerates the training of surrogate models representing the physical response of nanophotonic structures, reducing training time by at least an order of magnitude compared with conventional approaches. By replacing direct numerical solutions of partial differential equations (PDEs) with these surrogate models, the method achieves more than a two-order-of-magnitude reduction in simulation time. The study demonstrated a ten-layer metasurface consisting of 100 unit cells with a periodicity of 400 nm, capable of simultaneously focusing incident light at three discrete wavelengths (405 nm, 540 nm and 810 nm) onto three separate focal points. This multi-wavelength focusing performance highlights the versatility of the approach. Moreover, the authors emphasized that integrating active learning with surrogate modeling holds significant promise for accelerating the optimization of large-scale photonic devices, potentially enabling rapid design cycles and enhanced device functionalities. Similarly, as shown in Fig. 3(f), Seo et al. [117] proposed a deep learning–based image restoration framework for end-to-end meta-lens imaging. This approach effectively mitigates the severe chromatic and angular aberrations inherent in large-area broadband meta-lens, enabling aberration-free full-color imaging with mass-produced meta-lens of 10 mm in diameter. The proposed framework not only provides an effective strategy to overcome the intrinsic limitations of conventional optical systems but also establishes a foundation for the development of compact and high-efficiency imaging technologies. With neural network assistance, meta-lens imaging achieves resolutions comparable to ground-truth images.

2.2 Meta-grating

The grating is periodically incorporated into the metasurface structure, with metaatoms arranged in a periodic manner, thereby forming a meta-grating. This meta-grating allows precise control over the deflection of incident light into specific diffraction orders, enabling effective modulation of anomalous diffraction effects. Notably, this design not only improves diffraction efficiency but also achieves an abrupt phase shift covering the full 2π range at the interface. Furthermore, TO algorithms have been introduced into the design process of such meta-gratings. Jiang et al. [352] demonstrated that with training on visual data of periodic, topologically enhanced meta-gratings, generative neural networks can produce efficient, topologically intricate devices that can operate across a broad range of diffraction conditions. Subsequent iterative analysis further improves device robustness and efficiency, and these enhanced designs can serve as additional training data to refine the neural network model. As can be seen from Fig. 4(a), at short operating wavelengths, the efficiency histogram produced by the neural network method is similar to the efficiency histogram of the device with only iterative optimization. Nevertheless, GAN-based approaches have shown limited performance when compared to iterative optimization methods, indicating considerable room for improvement. Wen et al. [353] proposed a novel GAN training framework and network architecture specifically designed for meta-gratings with complex topologies. Their model facilitates the generation of highly efficient and robust free-form metasurfaces, achieving performance on par with or even exceeding that of gradient-based TO — all at a low computational cost. As illustrated in Fig. 4(b), the progressive growing GAN (PGGAN) integrates three major innovations: progressive network growth to improve the learning of local topological features; a self-attention mechanism for better modeling of global spatial dependencies within device patterns; and incremental expansion of the training dataset to more effectively represent efficient topological variations across the design space.

Inverse design algorithms play a crucial role in realizing high-performance, free-form nanophotonic devices. Chen et al. [354] revealed that redefining the design space itself can impose fundamental constraints on the development of inversely designed devices. The devices are first conceptualized in an unconstrained latent space and then translated into real physical space, where practical geometric restrictions can be efficiently enforced. Physical modifications to the device are mapped back into the latent representation via backpropagation. As a feasibility demonstration, the authors introduced a parameterization strategy to design efficient meta-gratings by applying strict minimum feature size constraints — using variable-based methods for localization and Global Topology Optimization Networks (GLOnets) for globalization. As illustrated in Fig. 4(c), which showcases the reparameterized design process of a meta-grating, this study utilizes robust, fixed-topology, parametric GLOnets to achieve global top-level optimization of a meta-grating under a 30 nm minimum feature size and 10 nm width variation constraint, ultimately reaching a relatively high efficiency exceeding 90%.

Inverse design algorithms are primarily used to simulate the optical responses corresponding to different combinations of structural parameters. In the literature, Inampudi et al. [355] proposed a method that utilizes ANNs to rapidly establish a numerical mapping relationship. This study employs a metasurface grating composed of periodic dielectric structures with arbitrary shapes. When incident plane waves illuminate the structure at normal incidence, the light is diffracted into 13 distinct orders. Thus, it is necessary to calculate the diffraction efficiency (DE) for each order generated by the arbitrary-shaped unit. Under the specific parameter combinations selected, the mean prediction error of the network remains below 0.04. As illustrated in Fig. 4(d), which shows a representative unit of the meta-grating, the approach demonstrates high predictive accuracy.

Recently, hybrid optimization strategies for customizing meta-gratings have garnered growing interest. As shown in Fig. 4(e), Elsawy et al. [356] employed a combination of statistical learning and evolutionary optimization techniques, integrated with a full-wave time-domain discontinuous Galerkin (DGtD) solver, to maximize the performance of phase-gradient metasurfaces. Their study focused on the optimal design of a GaN-based phase-gradient metasurface operating at visible wavelengths. Numerical results demonstrated that with only 150 full-wave simulations, arrays of both rectangular and cylindrical nanopillars achieved diffraction efficiencies exceeding 88% for TM polarization and over 85% for TE polarization, respectively.

2.3 Waveguide-based coupler

Free-space and integrated nanophotonic platforms have emerged as foundational platforms for photon manipulation, guidance, filtering, and transmission via subwavelength-engineered structures (see Refs. [357−367]). In response to the increasing demand for highly integrated optical circuits, a wide range of computational optimization techniques have been developed to reduce device footprints while enhancing overall performance. These include inverse design methods [368], adaptive gradient and adjoint-based optimization [167, 171, 222, 369−371], multi-objective optimization algorithms [372, 373], and multiskilled diffractive neural networks (MDNNs) [374].

The reverse-designed optical coupler represents a critical component within integrated photonic platforms. Du et al. [375] realized an ultra-compact multifunctional integrated photonic platform with a footprint of only 3 mm × 0.2 mm, integrating 86 micro-couplers and 91 phase shifters. The functionality of the platform was experimentally validated through quantum state evolution and photonic neural network tasks — such as handwritten digit recognition. As illustrated in Fig. 5(a), which shows a schematic of the platform, these experiments demonstrated high operational fidelity and achieved an on-chip test accuracy of 87%.

Similar to other nanophotonic interfaces, as shown in Fig. 5(b), Dory et al. [376] proposed an optimization-driven inverse design approach to overcome the fabrication constraints of diamond nanostructures and develop functional devices for diamond-based nanophotonic circuits. They demonstrated a grating coupler integrated with a diamond optical waveguide that enables interference between two nanobeam photonic crystal resonators via an inverse-designed waveguide splitter. This design achieved a simulated coupling efficiency of 95%, which marked a significant step forward in the development of integrated quantum photonic circuits. This design achieved a simulated coupling efficiency of 95%, marking a significant advancement in the development of integrated quantum photonic circuits.

As shown in Fig. 5(c), Gostimirovic et al. [377] developed an ANN model to accelerate the design of polarization-insensitive subwavelength grating (SWG) couplers on silicon photonic integrated platforms. The model optimizes SWG-based grating couplers by incorporating reverse design at the structural level to achieve either single- or dual-polarization operation. By mapping the nonlinear input–output behavior of the device into a learnable weight matrix, this method enables rapid prediction of device performance through straightforward linear algebraic operations and nonlinear activation functions. It achieves a computational speed approximately 1830 times faster than conventional numerical simulations while maintaining 93.2% accuracy compared to simulated results.

Beyond the aforementioned studies, additional research efforts have focused on optimizing other types of couplers. Jin et al. [378] proposed a scalable computational approach for designing highly efficient and compact on-chip devices capable of coupling light at multiple wavelengths. The designed device enables efficient broadband proximity coupling, covering six distinct wavelengths ranging from 560 nm to 1500 nm. As illustrated in Fig. 5(d), which shows a schematic of the waveguide coupler, this structure achieves high performance across the entire targeted spectrum.

2.4 Beam spiltter

A fundamental application of photonics lies in wavelength division multiplexing (WDM) [379−382], which increases the data transmission capacity of a single optical waveguide or fiber by employing multiple distinct wavelength channels. Piggott et al. [383] developed an inverse design algorithm capable of automatically generating various linear optical components. Users can specify target device functionalities, and the algorithm identifies structural designs that meet those requirements. Based on a local convex optimization framework, the method determines the distribution of dielectric constants and electric fields that satisfy both physical constraints and device performance targets. As a demonstration, a compact wavelength multiplexer, as shown in Fig. 6(a), was designed to separate 1300 nm and 1550 nm light from an input waveguide into two output waveguides. The device occupies an area of only 2.8 μm × 2.8 μm.

Building on this work, Piggott et al. [384] proposed an inverse design methodology that incorporates fabrication constraints. This approach was used to design several nanophotonic devices, including a 1 × 3 power splitter, a spatial-mode demultiplexer, a wavelength demultiplexer, and a two-port directional coupler. Experimental results of the 1 × 3 power splitter, as illustrated in Fig. 6(b), showed an insertion loss of 0.642 ± 0.057 dB and excellent power uniformity of 0.641 ± 0.054 dB, indicating low loss and high uniformity. The complex structural features of the device demonstrate the potential of inverse design for fabricating practical components in integrated photonic systems. Building on this foundation, the team subsequently proposed a novel one-dimensional photonic crystal split-beam [385] nanocavity, optimized through the integration of a deterministic design method with the hill-climbing algorithm, and experimentally demonstrated to achieve an ultra-high quality factor.

Su et al. [386] developed the Stanford photonic inverse design (SPIN) algorithm for nonlinear nanophotonic inverse design. By modularizing the inverse design workflow into interchangeable subcomponents, SPIN enables systematic and flexible exploration of diverse design strategies. The authors applied inverse design to generate hundreds of 3D wavelength demultiplexer designs and extensively investigated local minima in the design space, as illustrated in Fig. 6(c), thereby providing critical insights into the impact of initial conditions on device performance.

2.5 Light sensor

The ability to visually adapt and autonomously adjust responses to incoming light is a fundamental characteristic of intelligent biomimetic artificial vision systems. Various types of neuromorphic photocells have been developed, primarily based on low-dimensional materials [387], metal oxides [388, 389], ferroelectric materials [390], organic compounds [391−393], and perovskite photoactive materials [394, 395]. Thanks to their unique photosensitivity, these materials effectively utilize their photoconductive properties to detect and collect information from external optical stimuli. This functionality allows them to be integrated into advanced sensing and imaging systems that require precise and rapid optical signal acquisition.

As shown in Fig. 7(a), Lee et al. [396] developed a multidimensional photodetector based on an array of programmable light-responsive transistors (PLRTs) fabricated from black phosphorus (bP). The conductivity and photoresponsivity of these bP-PLRT devices can be dynamically programmed with 5-bit precision, both spatially and temporally, by modulating charge within the gate dielectric layer through combined electrical and optical control. This programmability enables the implementation of an onboard CNN. The sensor array is capable of simultaneously capturing optical images over a broad infrared spectral bandwidth and performing on-chip inference tasks with 92% image recognition accuracy, making it suitable for constructing more advanced visual perception neural networks.

As shown in Fig. 7(b), Zhu et al. [397] designed a novel flexible photoelectric sensor array comprising one thousand and twenty-four pixels, which utilizes a hybrid active material system of carbon nanotubes and perovskite quantum dots (CsPbBr₃-QDs) to enable efficient neuromorphic vision sensing. The device exhibits exceptional photosensitivity, with a measured responsivity of 5.1 × 10⁷ A/W and a specific detectivity of 2 × 10¹⁶ Jones. Furthermore, the sensor array was successfully trained via neuromorphic reinforcement learning using weak light pulses at an intensity of 1 μW/cm². These results underscore its potential to emulate the flexibility, complexity, and adaptability of biological visual processing, providing a promising platform for artificial neuromorphic vision systems. In a related effort, Luo et al. [374] proposed a multifunctional diffractive neuromorphic network implemented through an on-chip metasurface device, as shown in Fig. 7(c). By integrating a well-established complementary metal-oxide-semiconductor (CMOS) camera detector with the metasurface, they constructed a chip-scale architecture capable of processing images directly at the physical level. This system enables ultra-fast and highly energy-efficient image processing, with compelling applications in machine perception, autonomous vehicles, and advanced medical diagnostics.

Fiber-optic sensing, enhanced by artificial intelligence, now achieves higher accuracy and adaptability, enabling multifunctional signal detection and improved robustness against environmental disturbances. Wang et al. [398] proposed a pattern recognition strategy combining CNNs and LSTMs with phase-sensitive optical time-domain reflectometry for anomaly detection, achieving superior results on benchmark datasets. Expanding on this concept, Sabih and Vishwakarma [399] applied CNNs with bidirectional LSTMs to optical flow analysis, integrating domain knowledge to further improve classification accuracy. Leveraging these deep learning techniques in biomedical sensing, a mesh microbend fiber-optic sensor [400] was developed for perioperative monitoring of heart rate and respiratory rate, where an EMD-LSTM-CNN(ELC) model outperformed traditional FFT and WT methods, offering enhanced feature extraction and strong clinical potential.

2.6 Optical interference unit

Recently, due to the unique advantages of nano-optics such as parallel computing, low power consumption, and propagation at the speed of light, people are increasingly interested in using it to complete some specific machine learning tasks [401, 402]. The so-called artificial intelligence interference is mainly applied to photonic systems. The next two sections will introduce optical interference neural networks and optical diffraction neural networks respectively.

As shown in Fig. 8(a), Shen et al. [229] introduced a novel all-optical neural network architecture that harnesses the inherent advantages of optical computing to achieve a computational throughput at least two orders of magnitude higher than existing technologies, along with a three-order-of-magnitude improvement in power efficiency for conventional learning tasks. The architecture was experimentally realized using a cascaded array of Mach–Zehnder interferometers (MZIs), achieving a simulation accuracy of 91.7% in vowel recognition tasks, thereby demonstrating its practical applicability.

However, such optical networks often lack efficient training protocols and their physical implementation remains dependent on models simulated in electronic computers [403]. To facilitate on-chip training of ONNs, Hughes et al. [403] employed the adjoint variable method (AVM) to develop a photonic analogue of the backpropagation algorithm [404]. As shown in Fig. 8(b), they demonstrated the training process using a digitally simulated photonic ANN. In their work, the trained optical neural network was used to implement a logical XOR gate, successfully learning the gate function after approximately 400 iterations. The proposed method is also applicable to more complex computational tasks.

In addition to optical unit interfaces (OUI), a variety of functional computing units have been conceptualized and established as foundational components of ONNs [405−408]. As shown in Fig. 8(c), Estakhri et al. [407] proposed a metamaterial-based analog computing platform capable of solving partial differential equations through interaction with incident electromagnetic waves. When arbitrary input waveforms serve as functions corresponding to specific integral operators, the system yields a multifunctional electromagnetic field that represents the solution. The performance of the metastructure was verified through design, numerical simulation, and experimental validation at microwave frequencies. Experimental results indicated that the system reaches a steady-state resonant response within fewer than 300 cycles of a monochromatic input wave. Based on these observations, the authors estimated that the solution time reaches nanoseconds at microwave frequencies and picoseconds in the visible light regime. This approach offers a promising route toward compact, high-speed, and integrated on-chip analog computing systems.

Khoram et al. [409] employed a nanophotonic neural medium (NNM) to demonstrate and execute computer vision tasks. During training, a 20 × 20-pixel image is converted into a vector and encoded into the spatial intensity profile of the input light. Within the NNM, the nanostructure induces strong optical interference that modulates the propagation of light, directing it toward ten output ports corresponding to different classes. The port with the highest optical energy intensity is identified as the predicted category. The iterative training process, which utilizes mini-batch stochastic gradient descent, is illustrated in Fig. 8(d).

2.7 Optical diffractive neural network

Optical diffraction-based devices are also capable of performing a wide range of machine learning tasks [410]. Lin et al. [215] proposed a novel machine learning approach through an all-optical diffractive deep neural network (D²NN) architecture. This system leverages passive diffraction layers to computationally implement various functions through deep learning. A D²NN comprises a series of engineered diffractive layers, each designed to modulate incident light. Every point on these layers acts as an optical neuron, serving as a secondary source that re-radiates light. The properties of the emitted wave — both amplitude and phase — are determined by the interaction between the incoming optical field and the locally defined complex transmission or reflection coefficients. Through the cumulative modulation across multiple layers, D²NNs achieve task-specific optical transformations without relying on electronic processing. Connections between neurons in successive layers are established through secondary wave interference, modulated by prior layer diffraction patterns and local optical coefficients. The transmission or reflection coefficient at each point functions as a multiplicative bias term, which is optimized iteratively during training. As illustrated in Fig. 9(a), which depicts the network architecture and experimental outcomes, the cognitive capability of the D²NN was experimentally validated using a 3D-printed handwritten digit classifier. Numerical simulations of a five-layer D²NN achieved a classification accuracy of 91.75% on a test set of approximately 10,000 images. In contrast to earlier optoelectronic learning systems [229, 411−413], the D²NN framework provides a fully optical deep learning platform based on passive diffractive components, operating at the speed of light with high energy efficiency.

Several studies have investigated the use of ONNs operating in Fourier space for image processing applications. For example, Yan et al. [414] proposed a Fourier-space diffractive all-optical neural network (F-D²NN) situated at the Fourier plane within an optical imaging system. This configuration allows direct measurement of salient object features at the sensor. The study demonstrated the network’s capability for all-optical visual saliency detection through its application in cell segmentation. As shown in the right panel of Fig. 9(b), the saliency detection results clearly outline cell boundaries, accurately capturing their morphology and spatial distribution. The proposed architecture enables high-speed image processing and computer vision tasks at the speed of light, with its performance in high-accuracy visual saliency detection and object classification thoroughly validated through extensive numerical simulations.

More extensive neural network-like (NN-like) diffractive optical systems have been explored in recent literature, including reconfigurable diffractive processing units (DPUs), ensemble learning in diffractive optical networks [415], and broadband diffractive neural networks [416]. Zhou et al. [417] developed a reconfigurable DPU that can be programmed to implement various neural network architectures, facilitating large-scale neuromorphic optoelectronic computation. In their work, classification tasks were conducted on both handwritten digit datasets and human motion datasets. After adaptive training, the network achieved accuracies of 96.0% and 96.3%, respectively, resulting in highly accurate performance for rapid image and video recognition across multiple benchmark datasets. As illustrated in Fig. 9(c), the system demonstrates exceptional experimental accuracy. Furthermore, its computational capability outperforms that of state-of-the-art electronic computing platforms.

2.8 Other applications

The capability of AI algorithms to autonomously learn the nonlinear relationships between meta-devices and their electromagnetic responses has established them as an indispensable tool for designing high-performance optical devices. In addition to the optical components previously discussed, AI techniques have been extensively applied to a growing range of nanophotonic systems — including Fano resonators [418], noisy intermediate-scale quantum (NISQ) devices [419], photonic crystals [420], photon extractors [368], and particle accelerators [421] — significantly broadening the scope of nanodevice applications. In this section, we will further introduce applications in holographic imaging, nonlinear optical fibers, and optical information storage. These three examples illustrate the profound synergy between nanophotonic devices and artificial intelligence.

Optical holography enables the simultaneous acquisition of both phase and amplitude information of an object [69], and can even accurately reconstruct arbitrary three-dimensional vector field distributions of a wavefront [422]. Early research primarily focused on metasurface-based display and encryption of single holographic images [423, 424]. Recently, multiplexing has emerged as a superior alternative [425−430]. Wei et al. [67] established a spin-multiplexed metasurface inverse design platform based on a bidirectional deep neural network (Bi-DNN) model for rapidly generating high-pixel-density metasurfaces. As shown in Fig. 10(a), the meta-lens designed using this platform achieved focusing efficiencies of 54.06% and 50.49% in the two spin channels, respectively, with the near-field reconstructed holographic image closely matching the target image. Furthermore, by integrating the Bi-DNN model with advanced optimization algorithms, both electromagnetic response prediction and structural design can be enhanced, thereby supporting the development of high-performance photonic devices.

Teğin et al. [431] proposed and experimentally demonstrated an optical computing architecture termed the scalable optical learning operator (SOLO). This framework integrates a fixed nonlinear multimode fiber (MMF) mapping stage with a single-layer digital neural network to form a reconfigurable processing system. After training on a large dataset, the network can accurately identify the corresponding outputs, which are then recorded via a camera. The study conducted classification and prediction across multiple datasets, achieving an accuracy of 83% in diagnosing COVID-19 from lung X-ray images. As illustrated in Fig. 10(b), the experimental setup confirms the high stability of the SOLO framework.

Optical information storage [432−435] demonstrates broad application potential in meeting the demands of massive data storage owing to its high capacity, exceptional stability, and long lifespan. However, its further advancement remains challenged by the need to balance readout quality and error rates. In recent years, the integration of optical information storage with deep learning has opened new pathways toward high-fidelity data storage. Wang et al. [436] proposed a novel deep learning-based framework for optical storage, which converts information retrieved from a birefringence measurement system into binary data and feeds it into a neural network for classification, achieving nearly 100% accuracy. As illustrated in Fig. 10(c), experiments further confirmed that the deep learning model enables error-free information extraction even under external environmental interference. This experimental optical storage architecture lays the groundwork for realizing long-lifetime, highly stable optical storage systems capable of performing reliably in dynamic conditions, showcasing promising application prospects.

Nanophotonic represents an emerging frontier focused on manipulating light-matter interactions at the nanoscale, where the design of ultra-compact and high-performance devices remains a central challenge. Traditional empirical design methods are often computationally intensive and offer limited flexibility. In response, deep learning has enabled powerful inverse design strategies that begin with target optical functionalities and efficiently generate corresponding nanophotonic structures. This review has discussed how machine learning [163, 437−442] facilitates accurate modeling and optimization of nanostructures. Advanced algorithms, including federated learning [443, 444], transfer learning [445−447], and self-adaptive algorithm [448, 449] and distributed learning techniques [287, 450−453], have been successfully applied to the design of meta-lens, meta-gratings, beam splitters, and other key photonic components. In addition, hybrid optimization strategies [176, 284, 454, 455] and other optimization algorithms [456−463] have also garnered growing interest. These approaches significantly improve design efficiency, overcome local minima, and broaden solution space coverage.

3 Perspective and conclusion

Overall, AI has demonstrated considerable advantages in nanophotonic design, such as generating performance-targeted device designs within milliseconds and escaping local minima commonly encountered in traditional optimization processes. However, these methods also face several challenges: training neural networks requires large amounts of simulation data, and they remain sensitive to ill-posed inverse problems with multiple solutions, often resulting in poor convergence and limited coverage of the solution space. Future research may focus on physics-informed neural networks and reversible neural architectures. The former incorporates physical constraints such as Maxwell’s equations to reduce the demand for training data, while the latter addresses multi-solution issues through invertible network design. Combining AI with metaheuristic or gradient-based optimization is expected to further enhance design efficiency. Meanwhile, emerging photonic technologies such as ONNs, PICs and quantum photonics, provide new pathways toward all-optical AI hardware, enabling high-speed, low-power optical computing and real-time information processing. Although the integration of deep learning and nanophotonic device design still encounters bottlenecks — including limited algorithm generalizability and manufacturability constraints — breakthroughs in key technologies could unlock significant potential for developing novel photonic devices.

References

[1] A. Arbabi , Y. Horie , M. Bagheri , and A. Faraon , Dielectric metasurfaces for complete control of phase and polarization with subwavelength spatial resolution and high transmission, Nat. Nanotechnol. 10(11), 937 (2015)

[2] T. Cai , S. Tang , B. Zheng , G. Wang , W. Ji , C. Qian , Z. Wang , E. Li , and H. Chen , Ultrawideband chromatic aberration-free meta-mirrors, Adv. Photonics 3(1), 47 (2020)

[3] Z. Fan , C. Qian , Y. Jia , Z. Wang , Y. Ding , D. Wang , L. Tian , E. Li , T. Cai , B. Zheng , I. Kaminer , and H. Chen , Homeostatic neuro-metasurfaces for dynamic wireless channel management, Sci. Adv. 8(27), eabn7905 (2022)

[4] Y. Jia,C. Qian,Z. Fan,T. Cai,E. P. Li,H. Chen, A knowledge-inherited learning for intelligent metasurface design and assembly, Light Sci. Appl. 12(1), 82 (2023)

[5] D. Lin , P. Fan , E. Hasman , and M. L. Brongersma , Dielectric gradient metasurface optical elements, Science 345(6194), 298 (2014)

[6] X. Ni , A. V. Kildishev , and V. M. Shalaev , Metasurface holograms for visible light, Nat. Commun. 4(1), 2807 (2013)

[7] C. Qian , Z. Wang , H. Qian , T. Cai , B. Zheng , X. Lin , Y. Shen , I. Kaminer , E. Li , and H. Chen , Dynamic recognition and mirage using neuro-metamaterials, Nat. Commun. 13(1), 2694 (2022)

[8] N. Yu , P. Genevet , M. A. Kats , F. Aieta , J. P. Tetienne , F. Capasso , and Z. Gaburro , Light propagation with phase discontinuities: Generalized laws of reflection and refraction, Science 334(6054), 333 (2011)

[9] T. Hou , Y. Ge , S. Xue , and H. Chen , Curved anisotropic polaritons, Front. Phys. (Beijing) 19(3), 32201 (2024)

[10] X. Jing , X. Gui , P. Zhou , and Z. Hong , Physical explanation of Fabry–Pérot cavity for broadband bilayer metamaterials polarization converter, J. Lightwave Technol. 36(12), 2322 (2018)

[11] S. Li , Y. Fang , and J. Wang , Control of light–matter interactions in two-dimensional materials with nanoparticle-on-mirror structures, Opto-Electronic Science 3(7), 240011 (2024)

[12] M. Marishwari , S. Madhavamoorthi , N. Yogesh , and Z. Ouyang , Broadband terahertz multifunctional polarization conversion based on transmissive dipole ring meta-structures, Opt. Laser Technol. 174, 110644 (2024)

[13] R. Martín-Hernández,L. Grünewald,L. Sánchez-Tejerina, ., Optical magnetic field enhancement using ultrafast azimuthally polarized laser beams and tailored metallic nanoantennas, arXiv: 2024)

[14] G. Qin , Y. Dong , P. Zhang , J. Liu , C. Li , W. Huang , Z. Hong , C. Shen , and X. Jing , Broadband absorber coupled with anti-reflection inverted pyramid type all-dielectric microstructure, Opt. Lasers Eng. 160, 107288 (2023)

[15] Z. Refaeli,G. Marcus,Y. Shamir, A simple method for pulse contrast enhancement via self-focusing, High Power Laser Sci. Eng. 12, e18 (2024)

[16] B. Zeng , C. Li , H. Cheng , B. Fang , J. Lu , X. Ma , L. Wang , Z. Hong , and X. Jing , Anisotropic programmable metasurface beam splitter based on diode real-time control, Opt. Lasers Eng. 169, 107723 (2023)

[17] C. Zhang,K. Hu,J. Zhang,J. Cao,S. Zhang,H. F. Wang, Topological phase transition and edge states with tunable localization in the cyclic four‐mode optical system, Adv. Quantum Technol., 2400165 (2024)

[18] P. Mühlschlegel , H. J. Eisler , O. J. F. Martin , B. Hecht , and D. W. Pohl , Resonant optical antennas, Science 308(5728), 1607 (2005)

[19] A. M. Berhe , K. As’ham , I. Al-Ani , H. T. Hattori , and A. E. Miroshnichenko , Strong coupling and catenary field enhancement in the hybrid plasmonic metamaterial cavity and TMDC monolayers, Opto-Electronic Adv. 7(5), 230181 (2024)

[20] Y. Liu , Y. Tian , C. Li , B. Fang , J. Liu , Z. Hong , and X. Jing , All-dielectric double-layer honeycomb tunable metamaterial absorber with integrated gold nanoparticles, Photon. Res. 12(10), 2344 (2024)

[21] A. I. Kuznetsov , A. E. Miroshnichenko , M. L. Brongersma , Y. S. Kivshar , and B. Luk’yanchuk , Optically resonant dielectric nanostructures, Science 354(6314), aag2472 (2016)

[22] L. Jiang , B. Fang , Z. Yan , C. Li , J. Fu , H. Gan , Z. Hong , and X. Jing , Improvement of unidirectional scattering characteristics based on multiple nanospheres array, Microw. Opt. Technol. Lett. 62(6), 2405 (2020)

[23] X. Jing , Y. Xu , H. Gan , Y. He , and Z. Hong , High refractive index metamaterials by using higher order modes resonances of hollow cylindrical nanostructure in visible region, IEEE Access 7, 144945 (2019)

[24] H. Zeng , X. Cong , S. Wang , S. Gong , L. Huang , L. Wang , H. Liang , F. Lan , H. Cao , Z. Wang , W. Wang , S. Liang , Z. Feng , Z. Yang , Y. Zhang , and T. J. Cui , Ultrafast modulable 2DEG Huygens metasurface, Photon. Res. 12(5), 1004 (2024)

[25] C. Girard , Near fields in nanostructures, Rep. Prog. Phys. 68(8), 1883 (2005)

[26] M. Kauranen and A. V. Zayats , Nonlinear plasmonics, Nat. Photonics 6(11), 737 (2012)

[27] J. Jiang , B. Fang , C. Li , Z. Hong , and X. Jing , Improvement of terahertz beam modulation efficiency for baseless all-dielectric coded gratings, Photon. Res. 11(10), 1738 (2023)

[28] G. C. Francs and des , Plasmonic Purcell factor and coupling efficiency to surface plasmons: Implications for addressing and controlling optical nanosources, J. Opt. 18(9), 094005 (2016)

[29] B. Zeng , C. Li , B. Fang , Z. Hong , and X. Jing , Flexible tuning of multifocal holographic imaging based on electronically controlled metasurfaces, Photon. Res. 12(1), 61 (2024)

[30] X. Ni , Z. J. Wong , M. Mrejen , Y. Wang , and X. Zhang , An ultrathin invisibility skin cloak for visible light, Science 349(6254), 1310 (2015)

[31] C. Qian , B. Zheng , Y. Shen , L. Jing , E. Li , L. Shen , and H. Chen , Deep-learning-enabled self-adaptive microwave cloak without human intervention, Nat. Photonics 14(6), 383 (2020)

[32] D. Schurig , J. J. Mock , B. J. Justice , S. A. Cummer , J. B. Pendry , A. F. Starr , and D. R. Smith , Metamaterial electromagnetic cloak at microwave frequencies, Science 314(5801), 977 (2006)

[33] M. Deng , M. Cotrufo , J. Wang , J. Dong , Z. Ruan , A. Alù , and L. Chen , Broadband angular spectrum differentiation using dielectric metasurfaces, Nat. Commun. 15(1), 2237 (2024)

[34] P. Fu , Z. Xu , T. Zhou , H. Li , J. Wu , Q. Dai , and Y. Li , Reconfigurable metamaterial processing units that solve arbitrary linear calculus equations, Nat. Commun. 15(1), 6258 (2024)

[35] A. Silva , F. Monticone , G. Castaldi , V. Galdi , A. Alù , and N. Engheta , Performing mathematical operations with metamaterials, Science 343(6167), 160 (2014)

[36] H. Yang , X. Chen , E. Wu , P. Tang , L. Jing , F. Gao , H. Chen , and Z. Wang , Off‐resonance spin‐locked metasurfaces empowered by quasi‐bound states in the continuum for optical analog computing, Adv. Funct. Mater. 33(47), 2305110 (2023)

[37] E. Basar , M. Di Renzo , J. De Rosny , M. Debbah , M. S. Alouini , and R. Zhang , Wireless communications through reconfigurable intelligent surfaces, IEEE Access 7, 116753 (2019)

[38] M. Di Renzo , A. Zappone , M. Debbah , M. S. Alouini , C. Yuen , J. de Rosny , and S. Tretyakov , Smart radio environments empowered by reconfigurable intelligent surfaces: How it works, state of research, and the road ahead, IEEE J. Sel. Areas Comm. 38(11), 2450 (2020)

[39] C. Huang , A. Zappone , G. C. Alexandropoulos , M. Debbah , and C. Yuen , Reconfigurable intelligent surfaces for energy efficiency in wireless communication, IEEE Trans. Wirel. Commun. 18(8), 4157 (2019)

[40] W. Li , Q. Ma , C. Liu , Y. Zhang , X. Wu , J. Wang , S. Gao , T. Qiu , T. Liu , Q. Xiao , J. Wei , T. T. Gu , Z. Zhou , F. Li , Q. Cheng , L. Li , W. Tang , and T. J. Cui , Intelligent metasurface system for automatic tracking of moving targets and wireless communications based on computer vision, Nat. Commun. 14(1), 989 (2023)

[41] J. C. Liang,L. Zhang,Z. Luo,R. Z. Jiang,Z. W. Cheng,S. R. Wang,M. K. Sun,S. Jin,Q. Cheng,T. J. Cui, A filtering reconfigurable intelligent surface for interference-free wireless communications, Nat. Commun. 15(1), 3838 (2024)

[42] H. Takeshita , A. A. Fathnan , D. Nita , A. Nagata , S. Sugiura , and H. Wakatsuchi , Frequency-hopping wave engineering with metasurfaces, Nat. Commun. 15(1), 196 (2024)

[43] H. W. Tian , Y. L. Sun , X. G. Zhang , X. Li , Q. Zhu , C. Song , C. W. Qiu , T. J. Cui , and W. X. Jiang , Solar-powered light-modulated microwave programmable metasurface for sustainable wireless communications, Nat. Commun. 16(1), 2524 (2025)

[44] H. Zhan , M. Gu , Y. Tian , H. Feng , M. Zhu , H. Zhou , Y. Jin , Y. Tang , C. Li , B. Fang , Z. Hong , X. Jing , and L. Wang , Review for wireless communication technology based on digital encoding metasurfaces, Opto-Electron. Adv. 8(7), 240315 (2025)

[45] N. AbdelAll , M. Almokhtar , G. Khouqeer , . Rainbow trapping with engineered topological corner states and cavities in photonic crystals, Adv. Quantum Technol. 2024, 2400050 (2024)

[46] L. J. Roche , F. Betz , Y. Yang , I. Limame , C. W. Shih , S. Burger , and S. Reitzenstein , Numerical investigation of a coupled micropillar ‐ Waveguide system for integrated quantum photonic circuits, Adv. Quantum Technol. 7(12), 2400195 (2024)

[47] B. A. Taha , A. J. Addie , A. J. Haider , V. Chaudhary , R. Apsari , A. Kaushik , and N. Arsad , Exploring trends and opportunities in quantum‐enhanced advanced photonic illumination technologies, Adv. Quantum Technol. 7(3), 2300414 (2024)

[48] J. B. Pendry , Negative refraction makes a perfect lens, Phys. Rev. Lett. 85(18), 3966 (2000)

[49] D. R. Smith , W. J. Padilla , D. C. Vier , S. C. Nemat-Nasser , and S. Schultz , Composite medium with simultaneously negative permeability and permittivity, Phys. Rev. Lett. 84(18), 4184 (2000)

[50] X. He,F. Lin,F. Liu, ., 3D Dirac semimetals supported tunable terahertz BIC metamaterials, Nanophotonics 11(21), 4705 (2022)

[51] L. Jiang , B. Fang , Z. Yan , J. Fan , C. Qi , J. Liu , Y. He , C. Li , X. Jing , H. Gan , and Z. Hong , Terahertz high and near-zero refractive index metamaterials by double layer metal ring microstructure, Opt. Laser Technol. 123, 105949 (2020)

[52] S. Liu,W. Cao,S. Jiang,L. He,F. Lin,X. He, 3D Dirac semimetal supported tunable multi‐frequency terahertz metamaterial absorbers, Adv. Quantum Technol. 7(4), 2300386 (2024)

[53] J. Zhang , H. Zhang , and G. Zhang , Nanophononic metamaterials induced proximity effect in heat flux regulation, Front. Phys. (Beijing) 19(2), 23204 (2024)

[54] J. Zhao , X. Jing , W. Wang , Y. Tian , D. Zhu , and G. Shi , Steady method to retrieve effective electromagnetic parameters of bianisotropic metamaterials at one incident direction in the terahertz region, Opt. Laser Technol. 95, 56 (2017)

[55] W. L. Barnes , A. Dereux , and T. W. Ebbesen , Surface plasmon subwavelength optics, Nature 424(6950), 824 (2003)

[56] S. A. Maier , M. L. Brongersma , P. G. Kik , S. Meltzer , A. A. G. Requicha , and H. A. Atwater , Plasmonics - A route to nanoscale optical devices, Adv. Mater. 13(19), 1501 (2001)

[57] V. Istokskaia , B. Lefebvre , R. Versaci , D. B. Dreghici , D. Doria , F. Grepl , V. Olšovcová , F. Schillaci , S. Stanček , M. Tryus , A. Velyhan , D. Margarone , and L. Giuffrida , Real-time bremsstrahlung detector as a monitoring tool for laser–plasma proton acceleration, High Power Laser Sci. Eng. 12, e57 (2024)

[58] Y. Cheng , W. Cao , and X. He , Hybrid plasmonic waveguides with tunable ENZ phenomenon supported by 3D Dirac semimetals, Laser Photonics Rev. 18(9), 2400167 (2024)

[59] G. Wang,W. Cao,X. He, 3D Dirac semimetal elliptical fiber supported THz tunable hybrid plasmonic waveguides, IEEE J. Selected Topics Quantum Electron. 29(5), 1 (2023)

[60] S. Liu,W. Cao,J. Li, ., 3D Dirac semimetal metamaterial supported terahertz tunable dual-functional polarization converter, IEEE J. Selected Topics Quantum Electron. 31(5), 1 (2025)

[61] S. Y. Lin,J. G. Fleming,D. L. Hetherington,B. K. Smith,R. Biswas,K. M. Ho,M. M. Sigalas,W. Zubrzycki,S. R. Kurtz,J. Bur, A three-dimensional photonic crystal operating at infrared wavelengths, Nature 394(6690), 251 (1998)

[62] P. Russell , Photonic crystal fibers, Science 299(5605), 358 (2003)

[63] Z. Luo , Y. Ding , F. Peng , G. Wei , Y. Wang , and S. T. Wu , Ultracompact and high-efficiency liquid-crystal-on-silicon light engines for augmented reality glasses, Opto-Electron. Adv. 7(10), 240039 (2024)

[64] A. Ueno,J. Hu,S. An, AI for optical metasurface, npj Nanophotonics 1(1), 36 (2024)

[65] T. Phan , D. Sell , E. W. Wang , S. Doshay , K. Edee , J. Yang , and J. A. Fan , High-efficiency, large-area, topology-optimized metasurfaces, Light Sci. Appl. 8(1), 48 (2019)

[66] F. Léonard , A. S. Backer , E. J. Fuller , C. Teeter , and C. M. Vineyard , Co-design of free-space metasurface optical neuromorphic classifiers for high performance, ACS Photonics 8(7), 2103 (2021)

[67] H. Wei , X. He , and W. Cao , Spin-multiplexed metasurface inverse-design based on a bi-directional deep neural network for terahertz wavefront control, Optica 12(4), 505 (2025)

[68] R. Pestourie , Y. Mroueh , T. V. Nguyen , . Active learning of deep surrogates for PDEs: Application to metasurface design, npj Comput. Mater. 6(1), 164 (2020)

[69] G. Zheng , H. Mühlenbernd , M. Kenney , G. Li , T. Zentgraf , and S. Zhang , Metasurface holograms reaching 80% efficiency, Nat. Nanotechnol. 10(4), 308 (2015)

[70] E. Arbabi , S. M. Kamali , A. Arbabi , and A. Faraon , Full-Stokes imaging polarimetry using dielectric metasurfaces, ACS Photonics 5(8), 3132 (2018)

[71] H. Cheng , B. Fang , W. Huang , R. Li , C. Shen , L. Ke , X. Jing , C. Li , and Z. Hong , Far-field steering of beams based on dynamic encoding metasurface, Opt. Laser Technol. 156, 108520 (2022)

[72] L. Han , Q. Duan , J. Duan , S. Zhu , S. Chen , Y. Yin , and H. Chen , Unidirectional propagation of water waves near ancient Luoyang Bridge, Front. Phys. (Beijing) 19(3), 32206 (2024)

[73] Z. Hao , H. Chen , Y. Yin , C. W. Qiu , S. Zhu , and H. Chen , Efficient conversion of acoustic vortex using extremely anisotropic metasurface, Front. Phys. (Beijing) 19(4), 42202 (2024)

[74] M. W. Liang , D. Z. Lu , and Y. G. Ma , Vectorial optical fields manipulation via metasurfaces, Opto-Electron. Eng. 51(8), 240068 (2024)

[75] T. Nan , H. Zhao , J. Guo , X. Wang , H. Tian , and Y. Zhang , Generation of structured light beams with polarization variation along arbitrary spatial trajectories using tri-layer metasurfaces, Opto-Electron. Sci. 3(5), 230052 (2024)

[76] C. Peters,A. Forbes, Controlling the hidden parity in vectorial light with metasurfaces, Adv. Photonics 6(4) (2024)

[77] Q. Wang , C. Li , B. Fang , L. Ke , J. Lu , X. Ma , Z. Hong , and X. Jing , Flexible and independent control of asymmetric vortex beams based on spin multiplexing all dielectric metasurface, J. Lightwave Technol. 41(23), 7098 (2023)

[78] H. Yang , H. R. He , and Y. Q. Hu , Metasurface-empowered vector light field regulation, detection and application, Opto-Electron. Eng. 51(8), 240168 (2024)

[79] X. G. Zhang , Q. C. Hu , Q. Y. Li , Z. Huang , L. Xu , Y. Quan , Q. Zhu , B. Y. Wang , X. X. Chen , T. J. Cui , and W. X. Jiang , Anisotropic programmable metasurfaces with individually controllable 2-bit elements, Photon. Res. 12(11), 2660 (2024)

[80] Y. Zhang , Y. Chen , T. Wang , Q. Zhu , and M. Gu , Ultrahigh performance passive radiative cooling by hybrid polar dielectric metasurface thermal emitters, Opto-Electronic Advances 7(4), 230194 (2024)

[81] H. Zhao , X. Wang , S. Liu , and Y. Zhang , Highly efficient vectorial field manipulation using a transmitted tri-layer metasurface in the terahertz band, Opto-Electron. Adv. 6(2), 220012 (2023)

[82] Y. Chen , S. Zhang , Y. Tian , . Focus control of wide-angle metalens based on digitally encoded metasurface, Opto-Electron. Adv. 7(8), 240095 (2024)

[83] S. Baek , S. H. Park , D. Oh , K. Lee , S. Lee , H. Lim , T. Ha , H. S. Park , S. Zhang , L. Yang , B. Min , and T. T. Kim , Non-Hermitian chiral degeneracy of gated graphene metasurfaces, Light Sci. Appl. 12(1), 87 (2023)

[84] W. He , M. Tong , Z. Xu , Y. Hu , X. Cheng , and T. Jiang , Ultrafast all-optical terahertz modulation based on an inverse-designed metasurface, Photon. Res. 9(6), 1099 (2021)

[85] H. Lu,J. Zhao,B. Zheng, ., Eye accommodation-inspired neuro-metasurface focusing, Nat. Commun. 14(1) (2023)

[86] Q. Xu , X. Su , X. Zhang , L. Dong , L. Liu , Y. Shi , Q. Wang , M. Kang , A. Alù , S. Zhang , J. Han , and W. Zhang , Mechanically reprogrammable Pancharatnam–Berry metasurface for microwaves, Adv. Photonics 4(1), 016002 (2022)

[87] J. Sautter , I. Staude , M. Decker , E. Rusak , D. N. Neshev , I. Brener , and Y. S. Kivshar , Active tuning of all-dielectric metasurfaces, ACS Nano 9(4), 4308 (2015)

[88] Y. K. Srivastava , A. Chaturvedi , M. Manjappa , A. Kumar , G. Dayal , C. Kloc , and R. Singh , MoS₂ for ultrafast all‐optical switching and modulation of THz Fano metaphotonic devices, Adv. Opt. Mater. 5(23), 1700762 (2017)

[89] E. Wen , X. Yang , and D. F. Sievenpiper , Real-data-driven real-time reconfigurable microwave reflective surface, Nat. Commun. 14(1), 7736 (2023)

[90] J. Wu , T. Yuan , J. Liu , J. Qin , Z. Hong , J. Li , and Y. Du , Terahertz metamaterial sensor with ultra-high sensitivity and tunability based on photosensitive semiconductor GaAs, IEEE Sens. J. 22(16), 15961 (2022)

[91] X. Y. Wu , H. Y. Feng , F. Wan , M. Wei , C. Guo , L. Cai , F. Wu , Z. H. Jiang , L. Kang , W. Hong , and D. H. Werner , An ultrathin, fast‐response, large‐scale liquid‐crystal‐facilitated multi‐functional reconfigurable metasurface for comprehensive wavefront modulation, Adv. Mater. 36(26), 2402170 (2024)

[92] Y. Yao , R. Shankar , M. A. Kats , Y. Song , J. Kong , M. Loncar , and F. Capasso , Electrically tunable metasurface perfect absorbers for ultrathin mid-infrared optical modulators, Nano Lett. 14(11), 6526 (2014)

[93] X. Zhao , Y. Wang , J. Schalch , G. Duan , K. Cremin , J. Zhang , C. Chen , R. D. Averitt , and X. Zhang , Optically modulated ultra-broadband all-silicon metamaterial terahertz absorbers, ACS Photonics 6(4), 830 (2019)

[94] X. Zhu , C. Qian , E. Li , and H. Chen , Negative conductivity induced reconfigurable gain metasurfaces and their nonlinearity, Phys. Rev. Lett. 133(11), 113801 (2024)

[95] E. Arbabi,A. Arbabi,S. M. Kamali,Y. Horie,M. S. Faraji-Dana,A. Faraon, MEMS-tunable dielectric metasurface lens, Nat. Commun. 9(1), 812 (2018)

[96] H. S. Ee and R. Agarwal , Tunable metasurface and flat optical zoom lens on a stretchable substrate, Nano Lett. 16(4), 2818 (2016)

[97] T. Guo , Z. Zhang , Z. Lin , J. Tian , Y. Jin , J. Evans , Y. Xu , and S. He , Durable and programmable ultrafast nanophotonic matrix of spectral pixels, Nat. Nanotechnol. 19(11), 1635 (2024)

[98] P. Hosseini , C. D. Wright , and H. Bhaskaran , An optoelectronic framework enabled by low-dimensional phase-change films, Nature 511(7508), 206 (2014)

[99] R. Kaissner , J. Li , W. Lu , X. Li , F. Neubrech , J. Wang , and N. Liu , Electrochemically controlled metasurfaces with high-contrast switching at visible frequencies, Sci. Adv. 7(19), eabd9450 (2021)

[100] I. Kim , J. Jang , G. Kim , J. Lee , T. Badloe , J. Mun , and J. Rho , Pixelated bifunctional metasurface-driven dynamic vectorial holographic color prints for photonic security platform, Nat. Commun. 12(1), 3614 (2021)

[101] T. Lewi , H. A. Evans , N. A. Butakov , and J. A. Schuller , Ultrawide thermo-optic tuning of PbTe meta-atoms, Nano Lett. 17(6), 3940 (2017)

[102] Q. Li , J. van de Groep , A. K. White , J. H. Song , S. A. Longwell , P. M. Fordyce , S. R. Quake , P. G. Kik , and M. L. Brongersma , Metasurface optofluidics for dynamic control of light fields, Nat. Nanotechnol. 17(10), 1097 (2022)

[103] G. K. Shirmanesh , R. Sokhoyan , P. C. Wu , and H. A. Atwater , Electro-optically tunable multifunctional metasurfaces, ACS Nano 14(6), 6912 (2020)

[104] K. Sun , C. A. Riedel , Y. Wang , A. Urbani , M. Simeoni , S. Mengali , M. Zalkovskij , B. Bilenberg , C. H. de Groot , and O. L. Muskens , Metasurface optical solar reflectors using AZO transparent conducting oxides for radiative cooling of spacecraft, ACS Photonics 5(2), 495 (2018)

[105] S. Sun , W. Yang , C. Zhang , J. Jing , Y. Gao , X. Yu , Q. Song , and S. Xiao , Real-time tunable colors from microfluidic reconfigurable all-dielectric metasurfaces, ACS Nano 12(3), 2151 (2018)

[106] Z. Wang , L. Jing , K. Yao , Y. Yang , B. Zheng , C. M. Soukoulis , H. Chen , and Y. Liu , Origami‐based reconfigurable metamaterials for tunable chirality, Adv. Mater. 29(27), 1700412 (2017)

[107] X. Zhao , J. Schalch , J. Zhang , H. R. Seren , G. Duan , R. D. Averitt , and X. Zhang , Electromechanically tunable metasurface transmission waveplate at terahertz frequencies, Optica 5(3), 303 (2018)

[108] R. H. Xiong , X. Q. Peng , and J. S. Li , Graphene-metasurface for wide-incident-angle terahertz absorption, Front. Inf. Technol. Electron. Eng. 22(3), 334 (2020)

[109] E. A. Ash and G. Nicholls , Super-resolution aperture scanning microscope, Nature 237(5357), 510 (1972)

[110] E. Betzig , A. Harootunian , A. Lewis , and M. Isaacson , Near-field diffraction by a slit: Implications for superresolution microscopy, Appl. Opt. 25(12), 1890 (1986)

[111] L. Zhou , H. Lou , Z. Deng , X. Qin , J. Pan , Y. Di , C. Gu , D. Luo , and W. Li , Compact mid-infrared dual-comb spectrometer over 3–4 μm via intra-pulse difference frequency generation in LiNbO3 waveguides, High Power Laser Sci. Eng. 12, 12 (2024)

[112] B. Gallinet , J. Butet , and O. J. F. Martin , Numerical methods for nanophotonics: Standard problems and future challenges, Laser Photonics Rev. 9(6), 577 (2015)

[113] J. C. Zhang , M. K. Chen , Y. Fan , Q. Chen , S. Chen , J. Yao , X. Liu , S. Xiao , and D. P. Tsai , Miniature tunable Airy beam optical meta-device, Opto-Electronic Advances 7(2), 230171 (2024)

[114] S. Y. Xiang , Z. W. Song , and Y. H. Zhang , Progress in the research of optical neural networks, Guangdian Gongcheng 51(7), 30 (2024)

[115] S. Zhang , H. Li , C. Fan , Z. Zeng , C. Xiong , J. Wu , Z. Yan , D. Liu , and Q. Sun , Adaptive decentralized AI scheme for signal recognition of distributed sensor systems, Opto-Electronic Advances 7(12), 240119 (2024)

[116] X. Meng , C. Qin , X. Liang , G. Zhang , R. Chen , J. Hu , Z. Yang , J. Huo , L. Xiao , and S. Jia , Deep learning in two-dimensional materials: Characterization, prediction, and design, Front. Phys. (Beijing) 19(5), 53601 (2024)

[117] J. Seo,J. Jo,J. Kim,J. Kang,C. Kang,S. W. Moon,E. Lee,J. Hong,J. Rho,H. Chung, Deep-learning-driven end-to-end metalens imaging, Adv. Photonics 6(6) (2024)

[118] Z. Yu , M. Li , Z. Xing , H. Gao , Z. Liu , S. Pu , H. Mao , H. Cai , Q. Ma , W. Ren , J. Zhu , and C. Zhang , Genetic algorithm assisted meta-atom design for high-performance metasurface optics, Opto-Electronic Science 3(9), 240016 (2024)

[119] M. Sadman , S. Rahman , and A. Ozcan , Physics and artificial intelligence: Illuminating the future of optics and photonics, Adv. Photonics 6(5), 050500 (2024)

[120] B. Yang,X. Xiang,D. Xu,X. Wang,X. Yang, 3D palmprint recognition using shape index representation and fragile bits, Multimedia Tools Appl. 76(14), 15357 (2017)

[121] Y. Yang,X. Xu, A differential game model for closed-loop supply chain participants under carbon emission permits, Comput. Ind. Eng. 135, 1077 (2019)

[122] S. Jiang,H. Shi,W. Lin,H. C. Liu, A large group linguistic Z-DEMATEL approach for identifying key performance indicators in hospital performance management, Appl. Soft Comput. 86, 105900 (2020)

[123] J. Zhao,Z. Wang,F. Cao,D. Wang, A local learning algorithm for random weights networks, Knowl. Base. Syst. 74, 159 (2015)

[124] J. Chen,H. Cai,W. Wang, A new metaheuristic algorithm: Car tracking optimization algorithm., Soft Comput. 22(12), 3857 (2018)

[125] M. Hu,X. Feng,Z. Ji,K. Yan,S. Zhou, A novel computational approach for discord search with local recurrence rates in multivariate time series, Inf. Sci. 477, 220 (2019)

[126] D. Kong,J. Zhu,S. Wu,C. Duan,L. Lu,D. Chen, A novel IRBF-RVM model for diagnosis of atrial fibrillation, Comput. Methods Programs Biomed. 177, 183 (2019)

[127] H. Shi,L. Wang,X. Y. Li,H. C. Liu, A novel method for failure mode and effects analysis using fuzzy evidential reasoning and fuzzy Petri nets, J. Ambient Intell. Humaniz. Comput. 11(6), 2381 (2020)

[128] Z. Chai, A study of clustering on optimal fuzzy equivalence relations, Soft Comput. 27(3), 1415 (2023)

[129] X. Peng,D. Xu,J. Shen, A twin projection support vector machine for data regression, Neurocomputing 138, 131 (2014)

[130] X. F. Ding and H. C. Liu , An extended prospect theory–VIKOR approach for emergency decision making with 2-dimension uncertain linguistic information, Soft Comput. 23(22), 12139 (2019)

[131] X. G. Xu , Y. Xiong , D. H. Xu , and H. C. Liu , Bipolar fuzzy Petri nets for knowledge representation and acquisition considering non-cooperative behaviors, Int. J. Mach. Learn. Cybern. 11(10), 2297 (2020)

[132] H. Zou , Clustering algorithm and its application in data mining, Wirel. Pers. Commun. 110(1), 21 (2020)

[133] L. Hu , H. Wen , B. Wu , F. Pan , R. F. Liao , H. Song , J. Tang , and X. Wang , Cooperative Jamming for Physical Layer Security Enhancement in Internet of Things, IEEE Internet Things J. 5(1), 219 (2018)

[134] X. W. Jiang , T. H. Yan , J. J. Zhu , B. He , W. H. Li , H. P. Du , and S. S. Sun , Densely connected deep extreme learning machine algorithm, Cognit. Comput. 12(5), 979 (2020)

[135] Z. Hu , X. Wei , X. Han , G. Kou , H. Zhang , X. Liu , and Y. Bai , Density peaks clustering based on feature reduction and quasi-Monte Carlo, Sci. Program. 2022, 1 (2022)

[136] W. Zhu and Y. Peng , Elastic net regularized kernel non-negative matrix factorization algorithm for clustering guided image representation, Appl. Soft Comput. 97, 106774 (2020)

[137] X. Ding , L. Zhang , and H. Liu , Emergency decision making with extended axiomatic design approach under picture fuzzy environment, Expert Syst. 37(2), e12482 (2020)

[138] X. Y. Li , Y. Xiong , C. Y. Duan , and H. C. Liu , Failure mode and effect analysis using interval type-2 fuzzy sets and fuzzy Petri nets, J. Intell. Fuzzy Syst. 37(1), 693 (2019)

[139] J. Li , J. Hao , X. Sun , and Q. Feng , Forecasting China’s sovereign CDS with a decomposition reconstruction strategy, Appl. Soft Comput. 105, 107291 (2021)

[140] X. Wang , J. Wang , and K. Yan , Gait recognition based on Gabor wavelets and (2D)2PCA, Multimedia Tools Appl. 77(10), 12545 (2018)

[141] W. Wang , B. Lou , X. Li , X. Lou , N. Jin , and K. Yan , Intelligent maintenance frameworks of large-scale grid using genetic algorithm and K-Mediods clustering methods, World Wide Web (Bussum) 23(2), 1177 (2019)

[142] M. Luo and X. Zhou , Interval-valued quintuple implication principle of fuzzy reasoning, Int. J. Approx. Reason. 84, 23 (2017)

[143] H. Lu , Y. Meng , K. Yan , and Z. Gao , Kernel principal component analysis combining rotation forest method for linearly inseparable data, Cogn. Syst. Res. 53, 111 (2019)

[144] W. Zhu and B. Peng , Manifold-based aggregation clustering for unsupervised vehicle re-identification, Knowl. Base. Syst. 235, 107624 (2022)

[145] H. Wang , On extended progressive and iterative approximation for least squares fitting, Vis. Comput. 38(2), 591 (2022)

[146] J. Li , H. Wang , X. Wang , and Z. Li , Optimized sleep strategy based on clustering in dense heterogeneous networks, EURASIP J. Wirel. Commun. Netw. 2018(1), 290 (2018)

[147] K. Zhang and M. Luo , Outlier-robust extreme learning machine for regression problems, Neurocomputing 151, 1519 (2015)

[148] W. Li , S. Jiang , and Q. Jin , Overlap community detection using spectral algorithm based on node convergence degree, Future Gener. Comput. Syst. 79, 408 (2018)

[149] F. Cao , H. Hu , J. Lu , J. Zhao , Z. Zhou , and J. Wu , Pose and illumination variable face recognition via sparse representation and illumination dictionary, Knowl. Base. Syst. 107, 117 (2016)

[150] R. Xiang , H. Jiang , and Y. Ying , Recognition of clustered tomatoes based on binocular stereo vision, Comput. Electron. Agric. 106, 75 (2014)

[151] H. Han , J. Zhao , W. Zhai , Z. Xiong , D. Niyato , M. Di Renzo , Q. V. Pham , W. Lu , and K. Y. Lam , Reconfigurable Intelligent Surface Aided Power Control for Physical-Layer Broadcasting, IEEE Trans. Commun. 69(11), 7821 (2021)

[152] F. Cao , J. Chen , H. Ye , J. Zhao , and Z. Zhou , Recovering low-rank and sparse matrix based on the truncated nuclear norm, Neural Netw. 85, 10 (2017)

[153] Y. Liu and J. Wang , Research on post-match score mechanism of players based on artificial intelligence and clustering regression model, J. Intell. Fuzzy Syst. 39(4), 4869 (2020)

[154] M. Luo and K. Zhang , Robustness of full implication algorithms based on interval-valued fuzzy inference, Int. J. Approx. Reason. 62, 61 (2015)

[155] M. Luo and X. Zhou , Robustness of reverse triple I algorithms based on interval-valued fuzzy inference, Int. J. Approx. Reason. 66, 16 (2015)

[156] W. Zhu and B. Peng , Sparse and low-rank regularized deep subspace clustering, Knowl. Base. Syst. 204, 106199 (2020)

[157] X. Li,C. Zhu,C. Xu,J. Zhu,Y. Li,S. Wu, VR motion sickness recognition by using EEG rhythm energy ratio based on wavelet packet transform, Comput. Methods Programs Biomed. 188, 105266 (2020)

[158] D. Zhang and H. Zhou , Nonlinear bending and thermal post-buckling analysis of FGM beams resting on nonlinear elastic foundations, Comput. Model. Eng. Sci. 100(3), 201 (2014)

[159] L. Xie,Y. Yang,Z. Zhou,J. Zheng,M. Tao,Z. Man, Dynamic neural modeling of fatigue crack growth process in ductile alloys, Inf. Sci. 364–365, 167 (2016)

[160] S. Karumanchi , S. Mancini , A. Winter , and D. Yang , Classical capacities of quantum channels with environment assistance, Probl. Inf. Transm. 52(3), 214 (2016)

[161] Y. Kong , H. Lu , Y. Xue , and H. Xia , Terminal neural computing: Finite-time convergence and its applications, Neurocomputing 217, 133 (2016)

[162] F. Cao , X. Feng , and J. Zhao , Sparse representation for robust face recognition by dictionary decomposition, J. Vis. Commun. Image Represent. 46, 260 (2017)

[163] W. Wang and X. Liu , The selection of input weights of extreme learning machine: A sample structure preserving point of view, Neurocomputing 261, 28 (2017)

[164] D. Nie , J. Lin , and Q. Gao , Settling behavior of two particles with different densities in a vertical channel, Comput. Fluids 156, 353 (2017)

[165] J. Huang , Q. Wang , M. Fratarcangeli , K. Yan , and C. Pelachaud , Multi‐variate gaussian‐based inverse kinematics, Comput. Graph. Forum 36(8), 418 (2017)

[166] M. Luo and K. Zhou , Logical foundation of the quintuple implication inference methods, Int. J. Approx. Reason. 101, 1 (2018)

[167] W. Qian and F. Cao , Adaptive algorithms for low-rank and sparse matrix recovery with truncated nuclear norm, Int. J. Mach. Learn. Cybern. 10(6), 1341 (2019)

[168] X. Wang , T. Wang , J. Li , L. Shan , H. Cao , and Z. Li , Improved metric sorting for successive cancellation list decoding of polar codes, IEEE Commun. Lett. 23(7), 1123 (2019)

[169] C. Pan and W. Q. Yan , Object detection based on saturation of visual perception, Multimedia Tools Appl. 79(27-28), 19925 (2020)

[170] M. Luo and Y. Wang , Interval-valued fuzzy reasoning full implication algorithms based on the t-representable t-norm, Int. J. Approx. Reason. 122, 1 (2020)

[171] L. Wang and B. Wang , Construction of greenhouse environment temperature adaptive model based on parameter identification, Comput. Electron. Agric. 174, 105477 (2020)

[172] Z. D. Huang and H. D. Wang , On a progressive and iterative approximation method with memory for least square fitting, Comput. Aided Geom. Des. 82, 101931 (2020)

[173] C. Wen , W. Qian , Q. Zhang , and F. Cao , Algorithms of matrix recovery based on truncated Schatten p-norm, Int. J. Mach. Learn. Cybern. 12(5), 1557 (2021)

[174] Q. Li,Q. Y. Chen,Z. Liu,H. C. Liu, Public transport customer satisfaction evaluation using an extended thermodynamic method: a case study of Shanghai, China, Soft Comput. 25(16), 10901 (2021)

[175] P. Liu , H. He , T. Fu , H. Lu , A. Alelaiwi , and M. W. I. Wasi , Task offloading optimization of cruising UAV with fixed trajectory, Comput. Netw. 199, 108397 (2021)

[176] H. Lu,H. Gao,M. Ye,X. Wang, A hybrid ensemble algorithm combining AdaBoost and genetic algorithm for cancer classification with gene expression data, IEEE/ACM Trans. Comput. Biol. Bioinformatics 18(3), 863 (2021)

[177] X. Xu,Z. Yu,J. Zeng,W. Xiong,Y. Hu,G. Wang, A Bayesian belief-rule-based inference multivariate alarm system for nonlinear time-varying processes, Sci. China Inf. Sci. 64(10), 202203 (2021)

[178] G. Z. Wang,L. Xiong,H. C. Liu, A Bayesian inference method using Monte Carlo sampling for estimating the number of communities in bipartite networks, Sci. Program. 2019, 1 (2019)

[179] J. Li,M. Li,D. Wu,Q. Dai,H. Song, A Bayesian networks-based risk identification approach for software process risk: The context of Chinese trustworthy software, Int. J. Inf. Technol. Decis. Mak 15(6), 1391 (2016)

[180] Y. Xu , Y. Wang , T. Yan , Y. He , J. Wang , D. Gu , H. Du , and W. Li , Quality-related locally weighted soft sensing for non-stationary processes by a supervised Bayesian network with latent variables, Frontiers of Information Technology & Electronic Engineering 22(9), 1234 (2021)

[181] K. Dai,J. Zhao,F. Cao, A novel decorrelated neural network ensemble algorithm for face recognition, Knowl. Base. Syst. 89, 541 (2015)

[182] F. Cao,H. Ye,D. Wang, A probabilistic learning algorithm for robust modeling using neural networks with random weights, Inf. Sci. 313, 62 (2015)

[183] F. Cao , D. Wang , H. Zhu , and Y. Wang , An iterative learning algorithm for feedforward neural networks with random weights, Inf. Sci. 328, 546 (2016)

[184] X. Meng , J. Yan , H. Ye , and F. Cao , Construction and approximation for a class of feedforward neural networks with sigmoidal function, Int. J. Wavelets Multiresolution Inf. Process. 21(6), 2350028 (2023)

[185] F. Cao , K. Yao , and J. Liang , Deconvolutional neural network for image super-resolution, Neural Netw. 132, 394 (2020)

[186] H. Zhang , Y. Huang , B. Wang , and Z. Wang , Design and analysis of associative memories based on external inputs of delayed recurrent neural networks, Neurocomputing 136, 337 (2014)

[187] F. Cao , Y. Liu , and D. Wang , Efficient saliency detection using convolutional neural networks with feature selection, Inf. Sci. 456, 34 (2018)

[188] Y. Zhang , Y. Li , J. Sun , and J. Ji , Estimates on compressed neural networks regression, Neural Netw. 63, 10 (2015)

[189] J. Lu , J. Zhao , and F. Cao , Extended feed forward neural networks with random weights for face recognition, Neurocomputing 136, 96 (2014)

[190] J. J. Xiong , G. B. Zhang , J. X. Wang , and T. H. Yan , Improved sliding mode control for finite-time synchronization of nonidentical delayed recurrent neural networks, IEEE Trans. Neural Netw. Learn. Syst. 31(6), 2209 (2020)

[191] D. Lin , J. Wu , and J. Li , Less conservative stability condition for uncertain discrete-time recurrent neural networks with time-varying delays, Neurocomputing 173, 1578 (2016)

[192] J. Ren , Q. Zhang , F. Cao , C. Ding , and L. Wang , Modeling a stochastic age-structured capital system with Poisson jumps using neural networks, Inf. Sci. 516, 254 (2020)

[193] X. M. Zhang , Q. L. Han , Z. Wang , and B. L. Zhang , Neuronal state estimation for neural networks with two additive time-varying delay components, IEEE Trans. Cybern. 47(10), 3184 (2017)

[194] X. M. Zhang , Q. L. Han , X. Ge , and B. L. Zhang , Passivity analysis of delayed neural networks based on Lyapunov–Krasovskii functionals with delay-dependent matrices, IEEE Trans. Cybern. 50(3), 946 (2020)

[195] H. Cao , F. Cao , and D. Wang , Quantum artificial neural networks with applications, Inf. Sci. 290, 1 (2015)

[196] J. Miao , F. Cao , H. Ye , M. Li , and B. Yang , Revisiting graph neural networks from hybrid regularized graph signal reconstruction, Neural Netw. 157, 444 (2023)

[197] Z. Chen and F. Cao , Scattered data approximation by neural networks operators, Neurocomputing 190, 237 (2016)

[198] M. Pi , N. Jin , D. Chen , and B. Lou , Short-term solar irradiance prediction based on multichannel LSTM neural networks using edge-based IoT system, Wirel. Commun. Mob. Comput. 2022(1), 2372748 (2022)

[199] S. Lin and F. Cao , Simultaneous approximation by spherical neural networks, Neurocomputing 175, 348 (2016)

[200] J. Zhao , C. Chen , Z. Zhou , and F. Cao , Single image super-resolution based on adaptive convolutional sparse coding and convolutional neural networks, J. Vis. Commun. Image Represent. 58, 651 (2019)

[201] C. Zhu , B. Zheng , Z. He , M. Gao , C. Sun , and Z. Bao , State of health estimation of lithium-ion battery using time convolution memory neural network, Mob. Inf. Syst. 2021, 1 (2021)

[202] T. Hu , J. Zhao , R. Zheng , P. Wang , X. Li , and Q. Zhang , Ultrasonic based concrete defects identification via wavelet packet transform and GA-BP neural network, PeerJ Comput. Sci. 7, e635 (2021)

[203] Y. Zheng , L. He , X. Wu , and C. Pan , Self-training and multi-level adversarial network for domain adaptive remote sensing image segmentation, Neural Process. Lett. 55(8), 10613 (2023)

[204] W. Zhu , C. Chen , and B. Peng , Unified robust network embedding framework for community detection via extreme adversarial attacks, Inf. Sci. 643, 119200 (2023)

[205] W. Ma , F. Cheng , and Y. Liu , Deep-learning-enabled on-demand design of chiral metamaterials, ACS Nano 12(6), 6326 (2018)

[206] W. Ma , Z. Liu , Z. A. Kudyshev , A. Boltasseva , W. Cai , and Y. Liu , Deep learning for the design of photonic structures, Nat. Photonics 15(2), 77 (2021)

[207] Y. Xu , B. Xiong , W. Ma , and Y. Liu , Software-defined nanophotonic devices and systems empowered by machine learning, Prog. Quantum Electron. 89, 100469 (2023)

[208] Y. Rivenson , Y. Zhang , H. Günaydın , D. Teng , and A. Ozcan , Phase recovery and holographic image reconstruction using deep learning in neural networks, Light Sci. Appl. 7(2), 17141 (2017)

[209] G. Barbastathis , A. Ozcan , and G. Situ , On the use of deep learning for computational imaging, Optica 6(8), 921 (2019)

[210] C. S. Ho , N. Jean , C. A. Hogan , L. Blackmon , S. S. Jeffrey , M. Holodniy , N. Banaei , A. A. E. Saleh , S. Ermon , and J. Dionne , Rapid identification of pathogenic bacteria using Raman spectroscopy and deep learning, Nat. Commun. 10(1), 4927 (2019)

[211] C. Zuo , J. Qian , S. Feng , W. Yin , Y. Li , P. Fan , J. Han , K. Qian , and Q. Chen , Deep learning in optical metrology: A review, Light Sci. Appl. 11(1), 39 (2022)

[212] T. Fu , J. Zhang , R. Sun , Y. Huang , W. Xu , S. Yang , Z. Zhu , and H. Chen , Optical neural networks: Progress and challenges, Light Sci. Appl. 13(1), 263 (2024)

[213] Z. Huang , P. Wang , J. Liu , W. Xiong , Y. He , J. Xiao , H. Ye , Y. Li , S. Chen , and D. Fan , All-optical signal processing of vortex beams with diffractive deep neural networks, Phys. Rev. Appl. 15(1), 014037 (2021)

[214] C. Qian , X. Lin , X. Lin , J. Xu , Y. Sun , E. Li , B. Zhang , and H. Chen , Performing optical logic operations by a diffractive neural network, Light Sci. Appl. 9(1), 59 (2020)

[215] X. Lin , Y. Rivenson , N. T. Yardimci , M. Veli , Y. Luo , M. Jarrahi , and A. Ozcan , All-optical machine learning using diffractive deep neural networks, Science 361(6406), 1004 (2018)

[216] G. Wetzstein , A. Ozcan , S. Gigan , S. Fan , D. Englund , M. Soljačić , C. Denz , D. A. B. Miller , and D. Psaltis , Inference in artificial intelligence with deep optics and photonics, Nature 588(7836), 39 (2020)

[217] J. Feldmann , N. Youngblood , C. D. Wright , H. Bhaskaran , and W. H. P. Pernice , All-optical spiking neurosynaptic networks with self-learning capabilities, Nature 569(7755), 208 (2019)

[218] Z. Du,H. Ye,F. Cao, 3D mixed CNNs with edge-point feature learning, Knowl. Base. Syst. 221, 106985 (2021)

[219] S. Rao,G. Zou,S. Yang,S. Barmada, A feature selection and ensemble learning based methodology for transformer fault diagnosis, Appl. Soft Comput. 150, 111072 (2024)

[220] H. Lu,J. Chen,K. Yan,Q. Jin,Y. Xue,Z. Gao, A hybrid feature selection algorithm for gene expression data classification, Neurocomputing 256, 56 (2017)

[221] Z. Zhou,J. Zhao,F. Cao, A novel approach for fault diagnosis of induction motor with invariant character vectors, Inf. Sci. 281, 496 (2014)

[222] H. Ye,Y. Wang,F. Cao, A novel meta-learning framework: Multi-features adaptive aggregation method with information enhancer, Neural Netw. 144, 755 (2021)

[223] S. Zhao , Y. Wang , G. Deng , P. Yang , Z. Chen , and Y. Li , An intelligently adjusted carbon price forecasting approach based on breakpoints segmentation, feature selection and adaptive machine learning, Appl. Soft Comput. 149, 110948 (2023)

[224] C. Dai , C. Pan , and W. He , Feature extraction and fusion network for salient object detection, Multimedia Tools Appl. 81(23), 33955 (2022)

[225] X. Wang and J. Zhang , Gait feature extraction and gait classification using two-branch CNN, Multimedia Tools Appl. 79(3-4), 2917 (2020)

[226] Z. Gao , Z. Cheng , W. Diao , J. Zhang , and H. Lu , Identity authentication based on trajectory characteristics of mobile devices, J. Systems Archit. 112, 101857 (2021)

[227] X. Wang and W. Q. Yan , Non-local gait feature extraction and human identification, Multimedia Tools Appl. 80(4), 6065 (2021)

[228] G. Huang , X. Wang , X. Li , and Y. Wang , Spatiotemporal feature enhancement network for action recognition, Multimedia Tools Appl. 83(19), 57187 (2023)

[229] Y. Shen , N. C. Harris , S. Skirlo , M. Prabhu , T. Baehr-Jones , M. Hochberg , X. Sun , S. Zhao , H. Larochelle , D. Englund , and M. Soljačić , Deep Learning with Coherent Nanophotonic Circuits, Nat. Photonics 11(7), 441 (2017)

[230] X. F. Ding,H. C. Liu,H. Shi, A dynamic approach for emergency decision making based on prospect theory with interval-valued Pythagorean fuzzy linguistic variables, Comput. Ind. Eng. 131, 57 (2019)

[231] H. C. Liu,M. Y. Quan,Z. Li,Z. L. Wang, A new integrated MCDM model for sustainable supplier selection under interval-valued intuitionistic uncertain linguistic environment, Inf. Sci. 486, 254 (2019)

[232] R. Hu , B. Yang , H. Ye , F. Cao , C. Wen , and Q. Zhang , Decouple the object: Component-level semantic recognizer for point clouds classification, Knowl. Base. Syst. 248, 108887 (2022)

[233] L. X. Mao , R. Liu , X. Mou , and H. C. Liu , New approach for quality function deployment using linguistic z-numbers and EDAS method, Informatica (Vilnius) 32(3), 565 (2021)

[234] X. F. Han,J. S. Jin,M. J. Wang,W. Jiang,L. Gao,L. Xiao, A review of algorithms for filtering the 3D point cloud, Signal Process. Image Commun. 57, 103 (2017)

[235] J. Li , D. Mengu , Y. Luo , Y. Rivenson , and A. Ozcan , Class-specific differential detection in diffractive optical neural networks improves inference accuracy, Adv. Photonics 1(4), 1 (2019)

[236] R. Zhao,M. Luo,S. Li, A Dynamic Distance Measure of Picture Fuzzy Sets and Its Application, Symmetry (Basel) 13(3), 436 (2021)

[237] R. Zhao,L. Ma,S. Li,M. Luo, A multi-criteria three-way decision making method in a picture fuzzy probabilistic decision system, Cognit. Comput. 14(6), 1924 (2022)

[238] F. Cao,K. Li, A new method for image super-resolution with multi-channel constraints, Knowl. Base. Syst. 146, 118 (2018)

[239] M. Luo,Y. Zhang,L. Fu, A new similarity measure for picture fuzzy sets and its application to multi-attribute decision making, Informatica (Vilnius) 32(3), 543 (2021)

[240] S. Li,D. Li,R. Zhang,F. Cao, A novel autism spectrum disorder identification method: spectral graph network with brain-population graph structure joint learning, Int. J. Mach. Learn. Cybern. 15(4), 1517 (2024)

[241] F. Cao,C. Gao,H. Ye, A novel method for image segmentation: Two-stage decoding network with boundary attention, Int. J. Mach. Learn. Cybern. 13(5), 1461 (2022)

[242] B. Ke,H. Lu,C. You,W. Zhu,L. Xie,Y. Yao, A semi-supervised medical image classification method based on combined pseudo-labeling and distance metric consistency, Multimedia Tools Appl. 83(11), 33313 (2023)

[243] G. Huang , L. Yang , D. Zhang , X. Wang , and Y. Wang , Baggage image retrieval with attention-based network for security checks, Int. J. Pattern Recognit. Artif. Intell. 35(9), 2155009 (2021)

[244] J. Xiao , P. Liu , H. Lin , H. Fang , J. Xu , H. Shan , H. Hu , Y. Huang , and H. Lu , Blockchain and UAV-enabled signal source identification with edge computing and wireless signal-aerial image fusion, Wirel. Commun. Mob. Comput. 2022(1), 4009078 (2022)

[245] F. Cao and W. Guo , Cascaded dual-scale crossover network for hyperspectral image classification, Knowl. Base. Syst. 189, 105122 (2020)

[246] Z. Liang , H. Lu , R. Zhou , Y. Yao , and W. Zhu , CMFuse: Correlation-based multi-scale feature fusion network for the detection of COVID-19 from Chest X-ray images, Multimedia Tools Appl. 83(16), 49285 (2023)

[247] M. Ye , W. Zheng , H. Lu , X. Zeng , and Y. Qian , Cross-scene hyperspectral image classification based on DWT and manifold-constrained subspace learning, Int. J. Wavelets Multiresolution Inf. Process. 15(6), 1750062 (2017)

[248] M. Luo and G. Zhang , Divergence-based distance for picture fuzzy sets and its application to multi-attribute decision-making, Soft Comput. 28(1), 253 (2024)

[249] M. Ye , C. Wang , Z. Meng , F. Xiong , and Y. Qian , Domain-invariant attention network for transfer learning between cross-scene hyperspectral images, IET Comput. Vis. 17(7), 739 (2023)

[250] P. Gu , Y. Han , W. Gao , G. Xu , and J. Wu , Enhancing session-based social recommendation through item graph embedding and contextual friendship modeling, Neurocomputing 419, 190 (2021)

[251] J. Cao , K. Zhang , M. Luo , C. Yin , and X. Lai , Extreme learning machine and adaptive sparse representation for image classification, Neural Netw. 81, 91 (2016)

[252] M. Ye , Y. Xu , C. Ji , H. Chen , H. Lu , and Y. Qian , Feature selection for cross-scene hyperspectral image classification using cross-domain ReliefF, Int. J. Wavelets Multiresolution Inf. Process. 17(5), 1950039 (2019)

[253] Y. Song , S. Zhang , B. He , Q. Sha , Y. Shen , T. Yan , R. Nian , and A. Lendasse , Gaussian derivative models and ensemble extreme learning machine for texture image classification, Neurocomputing 277, 53 (2018)

[254] B. Shang,R. Chen,W. Yan,H. W. Yu, GPU accelerated volumetric lattice Boltzmann model for image-based hemodynamics in portal hypertension, Comput. Fluids 266, 106038 (2023)

[255] Z. Zheng,H. Wu,L. Lv,H. Ye,C. Zhang,G. Yu, ICCL: Independent and correlative correspondence learning for few-shot image classification, Knowl. Base. Syst. 266, 110412 (2023)

[256] R. Xiang , Image segmentation for whole tomato plant recognition at night, Comput. Electron. Agric. 154, 434 (2018)

[257] H. Liu , M. Shao , C. Wang , and F. Cao , Image super-resolution using a simple transformer without pretraining, Neural Process. Lett. 55(2), 1479 (2023)

[258] F. Cao , M. Cai , Y. Tan , and J. Zhao , Image super-resolution via adaptive ℓp(0<p<1) regularization and sparse representation, IEEE Trans. Neural Netw. Learn. Syst. 27(7), 1550 (2016)

[259] J. Zhao , H. Hu , and F. Cao , Image super-resolution via adaptive sparse representation, Knowl. Base. Syst. 124, 23 (2017)

[260] J. Zhao , T. Sun , and F. Cao , Image super‐resolution via adaptive sparse representation and self‐learning, IET Comput. Vis. 12(5), 753 (2018)

[261] H. Liu and F. Cao , Improved dual-scale residual network for image super-resolution, Neural Netw. 132, 84 (2020)

[262] H. Liu , Y. Li , R. Hong , Z. Li , M. Li , W. Pan , A. Glowacz , and H. He , Knowledge graph analysis and visualization of research trends on driver behavior, J. Intell. Fuzzy Syst. 38(1), 495 (2020)

[263] Z. Sun,J. Zhao,Z. Zhou,Q. Gao, L1 model-driven recursive multi-scale denoising network for image super-resolution, Knowl. Base. Syst. 225, 107115 (2021)

[264] B. He , Y. Song , Y. Zhu , Q. Sha , Y. Shen , T. Yan , R. Nian , and A. Lendasse , Local receptive fields based extreme learning machine with hybrid filter kernels for image classification, Multidimens. Syst. Signal Process. 30(3), 1149 (2019)

[265] C. Gao , H. Ye , F. Cao , C. Wen , Q. Zhang , and F. Zhang , Multiscale fused network with additive channel–spatial attention for image segmentation, Knowl. Base. Syst. 214, 106754 (2021)

[266] Y. Wen , L. A. Vese , K. Shi , Z. Guo , and J. Sun , Nonlocal adaptive biharmonic regularizer for image restoration, J. Math. Imaging Vis. 65(3), 453 (2023)

[267] Z. Wang , Y. Mao , S. Y. Hsieh , and J. Wu , On the g-good-neighbor connectivity of graphs, Theor. Comput. Sci. 804, 139 (2020)

[268] M. Luo and H. Long , Picture fuzzy geometric aggregation operators based on a trapezoidal fuzzy number and its application, Symmetry (Basel) 13(1), 119 (2021)

[269] Z. Zhou , W. Zhang , and J. Zhao , Robust visual tracking using discriminative sparse collaborative map, Int. J. Mach. Learn. Cybern. 10(11), 3201 (2019)

[270] Y. Liu , F. Cao , J. Zhao , and J. Chu , Segmentation of white blood cells image using adaptive location and iteration, IEEE J. Biomed. Health Inform. 21(6), 1644 (2017)

[271] M. Luo and W. Li , Some new similarity measures on picture fuzzy sets and their applications, Soft Comput. 27(10), 6049 (2023)

[272] Z. Chen , W. Fan , Z. Luo , and B. Guo , Soybean seed counting and broken seed recognition based on image sequence of falling seeds, Comput. Electron. Agric. 196, 106870 (2022)

[273] X. Li , C. Pan , L. He , and X. Li , Unsupervised domain adaptation for cross-domain histopathology image classification, Multimedia Tools Appl. 83(8), 23311 (2023)

[274] X. Wang , J. Li , L. Yang , H. Mi , and J. Y. Yu , Weakly-supervised learning for community detection based on graph convolution in attributed networks, Int. J. Mach. Learn. Cybern. 12(12), 3529 (2021)

[275] W. Liang , G. A. Tadesse , D. Ho , L. Fei-Fei , M. Zaharia , C. Zhang , and J. Zou , Advances, challenges and opportunities in creating data for trustworthy AI, Nat. Mach. Intell. 4(8), 669 (2022)

[276] D. Mengu , Y. Luo , Y. Rivenson , and A. Ozcan , Analysis of diffractive optical neural networks and their integration with electronic neural networks, IEEE J. Sel. Top. Quantum Electron. 26(1), 1 (2020)

[277] T. Wang , S. Y. Ma , L. G. Wright , T. Onodera , B. C. Richard , and P. L. McMahon , An optical neural network using less than 1 photon per multiplication, Nat. Commun. 13(1), 123 (2022)

[278] X. Meng,X. Lu,H. Ye,B. Yang,F. Cao, A new self-augment CNN for 3D point cloud classification and segmentation, Int. J. Mach. Learn. Cybern. 15(3), 807 (2024)

[279] X. Peng,J. Shen, A twin-hyperspheres support vector machine with automatic variable weights for data classification, Inf. Sci. 417, 216 (2017)

[280] W. Zheng , H. Tang , and Y. Qian , Collaborative work with linear classifier and extreme learning machine for fast text categorization, World Wide Web (Bussum) 18(2), 235 (2015)

[281] R. Zhang and X. Wang , Commodity classification in livestreaming marketing based on a conv-transformer network, Multimedia Tools Appl. 83(18), 54909 (2023)

[282] F. Cao , T. Dai , Y. Zhang , and Y. Tan , Compressed classification learning with Markov chain samples, Neural Netw. 50, 90 (2014)

[283] H. J. Lu , C. L. An , E. H. Zheng , . Dissimilarity based ensemble of extreme learning machine for gene expression data classification, Neurocomputing 128, 22 (2014)

[284] W. Wang , Y. Wang , J. Wang , X. Fang , and Y. He , Ensemble enhanced active learning mixture discriminant analysis model and its application for semi-supervised fault classification, Front. Inf. Technol. Electron. Eng. 23(12), 1814 (2022)

[285] X. Wang and K. Yan , Gait classification through CNN-based ensemble learning, Multimedia Tools Appl. 80(1), 1565 (2021)

[286] H. Lu , Y. Xu , M. Ye , K. Yan , Z. Gao , and Q. Jin , Learning misclassification costs for imbalanced classification on gene expression data, BMC Bioinformatics 20(S25), 681 (2019)

[287] G. Shi , C. Feng , W. Xu , L. Liao , and H. Huang , Penalized multiple distribution selection method for imbalanced data classification, Knowl. Base. Syst. 196, 105833 (2020)

[288] H. Lu , S. Wei , Z. Zhou , Y. Miao , and Y. Lu , Regularised extreme learning machine with misclassification cost and rejection cost for gene expression data classification, Int. J. Data Min. Bioinform. 12(3), 294 (2015)

[289] Y. Jin,H. Lu,Z. Li,Y. Wang, A cross-modal deep metric learning model for disease diagnosis based on chest X-ray images, Multimedia Tools Appl. 82(21), 33421 (2023)

[290] J. Zhao,Y. Lv,Z. Zhou,F. Cao, A novel deep learning algorithm for incomplete face recognition: Low-rank-recovery network, Neural Netw. 94, 115 (2017)

[291] L. Lei , B. Huang , M. Ye , F. Yao , and Y. Qian , Cross-domain residual deep NMF for transfer learning between different hyperspectral image scenes, Int. J. Wavelets Multiresolution Inf. Process. 21(02), 2250046 (2023)

[292] L. Lv , D. Bardou , Y. Liu , and P. Hu , Deep Autoencoder-like non-negative matrix factorization with graph regularized for link prediction in dynamic networks, Appl. Soft Comput. 148, 110832 (2023)

[293] F. Cao and W. Guo , Deep hybrid dilated residual networks for hyperspectral image classification, Neurocomputing 384, 170 (2020)

[294] Y. Jin , H. Lu , W. Zhu , and W. Huo , Deep learning based classification of multi-label chest X-ray images via dual-weighted metric loss, Comput. Biol. Med. 157, 106683 (2023)

[295] H. Zhou , Q. Liu , K. Yan , and Y. Du , Deep learning enhanced solar energy forecasting with AI‐driven IoT, Wirel. Commun. Mob. Comput. 2021(1), 9249387 (2021)

[296] K. Yan , X. Guo , Z. Ji , and X. Zhou , Deep transfer learning for cross-species plant disease diagnosis adapting mixed subdomains, IEEE/ACM Trans. Comput. Biol. Bioinformatics 20(4), 2555 (2023)

[297] Y. Zhou , J. Wu , S. Yao , Y. Xu , W. Zhao , Y. Tong , and Z. Zhou , DeepCIP: A multimodal deep learning method for the prediction of internal ribosome entry sites of circRNAs, Comput. Biol. Med. 164, 107288 (2023)

[298] J. Zhou , Y. Yin , J. Tang , Y. Xia , and J. Yin , Information transmission through parallel multi-task-based recognition of high-resolution multiplexed orbital angular momentum, Front. Phys. (Beijing) 19(5), 52202 (2024)

[299] C. Y. Shen,J. Li,Y. Li,T. Gan,L. Bai,M. Jarrahi,A. Ozcan, Multiplane quantitative phase imaging using a wavelength-multiplexed diffractive optical processor, Adv. Photonics 6(05) (2024)

[300] Q. Liu , X. Zhao , and K. Shi , The analysis of agricultural Internet of things product marketing by deep learning, J. Supercomput. 79(4), 4602 (2023)

[301] D. E. Rumelhart , G. E. Hinton , and R. J. Williams , Learning representations by back-propagating errors, Nature 323(6088), 533 (1986)

[302] P. J. Werbos , Backpropagation through time: what it does and how to do it, Proc. IEEE 78(10), 1550 (1990)

[303] X. Jin,X. Li,Y. Feng,M. Wen,C. Huang,Y. Yao, RIS-based opposite phase index modulation and its DNN-assisted joint sequence detection, IEEE Wirel. Commun. Lett. 12(9), 1518 (2023)

[304] N. Wang , W. Yan , Y. Qu , S. Ma , S. Z. Li , and M. Qiu , Intelligent designs in nanophotonics: From optimization towards inverse creation, PhotoniX 2(1), 22 (2021)

[305] P. R. Wiecha , A. Arbouet , C. Girard , and O. L. Muskens , Deep learning in nano-photonics: Inverse design and beyond, Photon. Res. 9(5), B182 (2021)

[306] K. Yao , R. Unni , and Y. Zheng , Intelligent nanophotonics: merging photonics and artificial intelligence at the nanoscale, Nanophotonics 8(3), 339 (2019)

[307] M. Abadi,P. Barham,J. Chen, ., TensorFlow: A system for large-scale machine learning[C]//Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation. USA: USENIX Association, 2016, pp 265–283

[308] K. He,X. Zhang,S. Ren, ., Deep residual learning for image recognition tensor flow: A system for large-scale machine learning, in: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2016, pp 770–778

[309] M. Bertolotti, The History of the Laser, CRC Press, 2004

[310] G. Schaack and A. Bela , Lengyel: Lasers, generation of light by stimulated emission, Ber. Bunsenges. Phys. Chem 67(6), 614 (1963)

[311] G. Brambilla , V. Finazzi , and D. J. Richardson , Ultra-low-loss optical fiber nanotapers, Opt. Express 12(10), 2258 (2004)

[312] Y. Yamamoto , Low-loss optical fibers, Review of Laser Engineering 40(6), 411 (2012)

[313] P. Kaushik,D. P. Singh,S. Rajpoot, Fibre Optic Communication in 21st Century, 2020 International Conference on Intelligent Engineering and Management (ICIEM), p. 125 (2020)

[314] T. Otani,K. Goto,H. Abe,M. Tanaka,H. Yamamoto,H. Wakabayashi, 5.3 Gbit/s 11300 km data transmission using actual submarine cables and repeaters, Electron. Lett. 31(5), 380 (1995)

[315] D. B. Keck , Observation of externally controlled mode coupling in optical waveguides, Proc. IEEE 62(5), 649 (1974)

[316] E. L. Chinnock , L. G. Cohen , W. S. Holden , R. D. Standley , and D. B. Keck , The length dependence of pulse spreading in the CGW-Bell-10 optical fiber, Proc. IEEE 61(10), 1499 (1973)

[317] C. A. Burrus , E. L. Chinnock , D. Gloge , W. S. Holden , Tingye Li , R. D. Standley , and D. B. Keck , Pulse dispersion and refractive-index profiles of some low-noise multimode optical fibers, Proc. IEEE 61(10), 1498 (1973)

[318] S. Zlatanovic , J. S. Park , S. Moro , J. M. C. Boggio , I. B. Divliansky , N. Alic , S. Mookherjea , and S. Radic , Mid-infrared wavelength conversion in silicon waveguides using ultracompact telecom-band-derived pump source, Nat. Photonics 4(8), 561 (2010)

[319] K. Uesaka , K. K. Y. Wong , M. E. Marhic , and L. G. Kazovsky , Wavelength exchange in a highly nonlinear dispersion-shifted fiber: Theory and experiments, IEEE J. Sel. Top. Quantum Electron. 8(3), 560 (2002)

[320] A. C. Turner , M. A. Foster , A. L. Gaeta , and M. Lipson , Ultra-low power parametric frequency conversion in a silicon microring resonator, Opt. Express 16(7), 4881 (2008)

[321] Q. Li , M. Davanço , and K. Srinivasan , Efficient and low-noise single-photon-level frequency conversion interfaces using silicon nanophotonics, Nat. Photonics 10(6), 406 (2016)

[322] S. Lefrancois , A. S. Clark , and B. J. Eggleton , Optimizing optical Bragg scattering for single-photon frequency conversion, Phys. Rev. A 91(1), 013837 (2015)

[323] C. Langrock , E. Diamanti , R. V. Roussev , Y. Yamamoto , M. M. Fejer , and H. Takesue , Highly efficient single-photon detection at communication wavelengths by use of upconversion in reverse-proton-exchanged periodically poled LiNbO3 waveguides, Opt. Lett. 30(13), 1725 (2005)

[324] A. H. Gnauck , R. M. Jopson , C. J. McKinstrie , J. C. Centanni , and S. Radic , Demonstration of low-noise frequency conversion by Bragg Scattering in a Fiber, Opt. Express 14(20), 8989 (2006)

[325] L. Zhuang , C. G. H. Roeloffzen , M. Hoekman , K. J. Boller , and A. J. Lowery , Programmable photonic signal processor chip for radiofrequency applications, Optica 2(10), 854 (2015)

[326] A. Ribeiro , A. Ruocco , L. Vanacker , and W. Bogaerts , Demonstration of a 4 × 4-port universal linear circuit, Optica 3(12), 1348 (2016)

[327] J. Notaros , J. Mower , M. Heuck , C. Lupo , N. C. Harris , G. R. Steinbrecher , D. Bunandar , T. Baehr-Jones , M. Hochberg , S. Lloyd , and D. Englund , Programmable dispersion on a photonic integrated circuit for classical and quantum applications, Opt. Express 25(18), 21275 (2017)

[328] P. L. Mennea , W. R. Clements , D. H. Smith , J. C. Gates , B. J. Metcalf , R. H. S. Bannerman , R. Burgwal , J. J. Renema , W. S. Kolthammer , I. A. Walmsley , and P. G. R. Smith , Modular linear optical circuits, Optica 5(9), 1087 (2018)

[329] Y. Wu , Q. Yang , B. Shen , Y. Tao , X. Zhang , Z. Tao , L. Xing , Z. Ge , T. Li , B. Bai , H. Shu , and X. Wang , Multifunctional mixed analog/digital signal processor based on integrated photonics, Opto-Electronic Science 3(8), 240012 (2024)

[330] J. Shi , C. Yang , Y. Chen , T. Wang , H. Yu , J. Cao , Z. Geng , Z. Wang , H. Wen , H. Tan , Y. Zhang , D. Jiang , D. Wu , Y. Xu , H. Ni , and Z. Niu , Precise mode control of mid-infrared high-power laser diodes using on-chip advanced sawtooth waveguide designs, High Power Laser Sci. Eng. 12, e42 (2024)

[331] H. Han,L. Fang,W. Lu,W. Zhai,Y. Li,J. Zhao, A GCICA grant-free random access scheme for M2M communications in crowded massive MIMO systems, IEEE Internet Things J. 9(8), 6032 (2022)

[332] Y. Tan,F. Cao,M. Cai, A new system of face recognition: Using fuzziness and sparsity, Int. J. Uncertain. Fuzziness Knowl. Based Syst. 23(6), 829 (2015)

[333] J. Li,S. Chen,L. Duan,L. Niu, A PR-quadtree based multi-dimensional indexing for complex query in a cloud system, Cluster Computing 20(4), 2931 (2017)

[334] X. Wang and K. Yan , Automatic color correction for multi-projector display systems, Multimedia Tools Appl. 77(11), 13115 (2018)

[335] Z. Gao , H. Guo , Y. Xie , Y. Luo , H. Lu , and K. Yan , ChildGuard: A child-safety monitoring system, IEEE Multimed. 24(4), 48 (2017)

[336] X. Jin , Q. Wu , Q. Li , W. C. Zhai , and N. Jin , Design of power allocation for APSK non-coherent spatial modulation system, Mob. Netw. Appl. 25(5), 1715 (2020)

[337] D. Zhang , J. Xu , M. Sun , and Z. Xiang , High-density crowd behaviors segmentation based on dynamical systems, Multimedia Syst. 23(5), 599 (2017)

[338] X. Wang and K. Yan , Immersive human–computer interactive virtual environment using large-scale display system, Future Gener. Comput. Syst. 96, 649 (2019)

[339] Z. Yu and H. Yu , Multicultural integration of urban international communities in the innovation ecosystem, Sci. Program. 2022, 1 (2022)

[340] X. Cui , J. Chen , B. Wang , and S. Xu , Off-policy algorithm based hierarchical optimal control for completely unknown dynamic systems, Neurocomputing 488, 669 (2022)

[341] H. Zheng,D. Liu,Y. Liu, RETRACTED: Design and research on automatic recognition system of sports dance movement based on computer vision and parallel computing, Microprocess. Microsyst. 80, 103648 (2021)

[342] Y. Li , Z. Zuo , and J. Pan , Sensor-based fall detection using a combination model of a temporal convolutional network and a gated recurrent unit, Future Gener. Comput. Syst. 139, 53 (2023)

[343] Z. Ying , Y. Wang , Y. He , and J. Wang , Virtual sensing techniques for nonlinear dynamic processes using weighted probability dynamic dual-latent variable model and its industrial applications, Knowl. Base. Syst. 235, 107642 (2022)

[344] S. Rane , S. Prabhu , and D. R. Chowdhury , Physics and applications of terahertz metagratings, Opto-Electronic Science 3(11), 230049 (2024)

[345] X. Jing , X. Tang , Y. Tian , Z. Kong , C. Li , C. Shen , and Z. Hong , Enhancement of diffraction efficiency based on the addition principle of coded digital gratings, J. Lightwave Technol. 40(1), 136 (2022)

[346] Y. Fan , Y. Xu , M. Qiu , W. Jin , L. Zhang , E. Y. Lam , D. P. Tsai , and D. Lei , Phase-controlled metasurface design via optimized genetic algorithm, Nanophotonics 9(12), 3931 (2020)

[347] E. Bayati , R. Pestourie , S. Colburn , Z. Lin , S. G. Johnson , and A. Majumdar , Inverse designed extended depth of focus meta-optics for broadband imaging in the visible, Nanophotonics 11(11), 2531 (2022)

[348] R. E. Christiansen , Z. Lin , C. Roques-Carmes , Y. Salamin , S. E. Kooi , J. D. Joannopoulos , M. Soljačić , and S. G. Johnson , Fullwave Maxwell inverse design of axisymmetric, tunable, and multi-scale multi-wavelength metalenses, Opt. Express 28(23), 33854 (2020)

[349] R. E. Christiansen and O. Sigmund , Compact 200 line MATLAB code for inverse design in photonics by topology optimization: Tutorial, J. Opt. Soc. Am. B 38(2), 510 (2021)

[350] Z. Lin , C. Roques-Carmes , R. E. Christiansen , M. Soljačić , and S. G. Johnson , Computational inverse design for ultra-compact single-piece metalenses free of chromatic and angular aberration, Appl. Phys. Lett. 118(4), 041104 (2021)

[351] Z. Lin and S. G. Johnson , Overlapping domains for topology optimization of large-area metasurfaces, Opt. Express 27(22), 32445 (2019)

[352] J. Jiang , D. Sell , S. Hoyer , J. Hickey , J. Yang , and J. A. Fan , Free-form diffractive metagrating design based on generative adversarial networks, ACS Nano 13(8), 8872 (2019)

[353] F. Wen , J. Jiang , and J. A. Fan , Robust freeform metasurface design based on progressively growing generative networks, ACS Photonics 7(8), 2098 (2020)

[354] M. Chen , J. Jiang , and J. A. Fan , Design space reparameterization enforces hard geometric constraints in inverse-designed nanophotonic devices, ACS Photonics 7(11), 3141 (2020)

[355] S. Inampudi and H. Mosallaei , Neural network based design of metagratings, Appl. Phys. Lett. 112(24), 241102 (2018)

[356] M. M. R. Elsawy , S. Lanteri , R. Duvigneau , G. Brière , M. S. Mohamed , and P. Genevet , Global optimization of metasurface designs using statistical learning methods, Sci. Rep. 9(1), 17918 (2019)

[357] W. Wang,X. F. Jing,J. Zhao,Y. Li,Y. Tian, Improvement of accuracy of simple methods for design and analysis of a blazed phase grating microstructure, Opt. Appl., doi: 10.5277/OA170202 (2017)

[358] M. Ullah and S. Mikki , Optical nonreciprocity in a multimode cavity optomechanical system controlled by dynamic Casimir force, Adv. Quantum Technol. 7(7), 2400007 (2024)

[359] X. Zou , Y. Zhang , R. Lin , G. Gong , S. Wang , S. Zhu , and Z. Wang , Pixel-level Bayer-type colour router based on metasurfaces, Nat. Commun. 13(1), 3288 (2022)

[360] A. Zaidi , N. A. Rubin , M. L. Meretska , L. W. Li , A. H. Dorrah , J. S. Park , and F. Capasso , Metasurface-enabled single-shot and complete Mueller matrix imaging, Nat. Photonics 18(7), 704 (2024)

[361] N. Yu and F. Capasso , Flat optics with designer metasurfaces, Nat. Mater. 13(2), 139 (2014)

[362] S. Shrestha , A. C. Overvig , M. Lu , A. Stein , and N. Yu , Broadband achromatic dielectric metalenses, Light Sci. Appl. 7(1), 85 (2018)

[363] S. M. Kamali,E. Arbabi,A. Arbabi, ., A review of dielectric optical metasurfaces for wavefront control, Nanophotonics 7(6), 1041 (2018)

[364] X. Hua , Y. Wang , S. Wang , X. Zou , Y. Zhou , L. Li , F. Yan , X. Cao , S. Xiao , D. P. Tsai , J. Han , Z. Wang , and S. Zhu , Ultra-compact snapshot spectral light-field imaging, Nat. Commun. 13(1), 2732 (2022)

[365] T. Dai , Y. Shao , C. Mao , Y. Wu , S. Azzouz , Y. Zhou , and J. A. Fan , Shaping freeform nanophotonic devices with geometric neural parameterization, npj Comput. Mater. 11(1), 259 (2025)

[366] A. Chen and F. Monticone , Dielectric nonlocal metasurfaces for fully solid-state ultrathin optical systems, ACS Photonics 8(5), 1439 (2021)

[367] C. Guo,H. Wang,S. Fan, Squeeze Free Space Using Nonlocal Flat Optics[C]. Washington, DC: Optica Publishing Group, 2021: FM4B. 4.

[368] S. Chakravarthi , P. Chao , C. Pederson , S. Molesky , A. Ivanov , K. Hestroffer , F. Hatami , A. W. Rodriguez , and K. M. C. Fu , Inverse-designed photon extractors for optically addressable defect qubits, Optica 7(12), 1805 (2020)

[369] F. Cao,L. Zhu,H. Ye,C. Wen,Q. Zhang, A new method for point cloud registration: Adaptive relation-oriented convolution and recurrent correspondence-walk, Knowl. Base. Syst. 284, 111280 (2024)

[370] H. Wang,H. Ye,B. Yang,F. Cao, A novel method for point cloud completion: Adaptive region shape fusion network, Knowl. Base. Syst. 255, 109770 (2022)

[371] X. Wang , F. Cao , and W. Wang , Adaptive sparse and dense hybrid representation with nonconvex optimization, Front. Comput. Sci. 14(4), 144306 (2020)

[372] J. Huang , J. Yang , D. Chen , W. Bai , J. Han , Z. Zhang , J. Zhang , X. He , Y. Han , and L. Liang , Implementation of on-chip multi-channel focusing wavelength demultiplexer with regularized digital metamaterials, Nanophotonics 9(1), 159 (2020)

[373] T. Zhou , S. Zhang , D. Zhang , V. Chan , S. Yang , and M. Chen , Customer-oriented multi-objective optimization on a novel collaborative multi-heterogeneous-depot electric vehicle routing problem with mixed time windows, J. Intell. Fuzzy Syst. 44(3), 3787 (2023)

[374] X. Luo , Y. Hu , X. Ou , X. Li , J. Lai , N. Liu , X. Cheng , A. Pan , and H. Duan , Metasurface-enabled on-chip multiplexed diffractive neural networks in the visible, Light Sci. Appl. 11(1), 158 (2022)

[375] Z. Du , K. Liao , T. Dai , Y. Wang , J. Gao , H. Huang , H. Qi , Y. Li , X. Wang , X. Su , X. Wang , Y. Yang , C. Lu , X. Hu , and Q. Gong , Ultracompact and multifunctional integrated photonic platform, Sci. Adv. 10(25), eadm7569 (2024)

[376] C. Dory , D. Vercruysse , K. Y. Yang , N. V. Sapra , A. E. Rugar , S. Sun , D. M. Lukin , A. Y. Piggott , J. L. Zhang , M. Radulaski , K. G. Lagoudakis , L. Su , and J. Vučković , Inverse-designed diamond photonics, Nat. Commun. 10(1), 3309 (2019)

[377] D. Gostimirovic and W. N. Ye , An open-source artificial neural network model for polarization-insensitive silicon-on-insulator subwavelength grating couplers, IEEE J. Sel. Top. Quantum Electron. 25(3), 1 (2019)

[378] W. Jin , S. Molesky , Z. Lin , K. M. C. Fu , and A. W. Rodriguez , Inverse design of compact multimode cavity couplers, Opt. Express 26(20), 26713 (2018)

[379] H. Ren , X. Li , Q. Zhang , and M. Gu , On-chip noninterference angular momentum multiplexing of broadband light, Science 352(6287), 805 (2016)

[380] Y. Yuan,Y. Peng,W. V. Sorin,S. Cheung,Z. Huang,D. Liang,M. Fiorentino,R. G. Beausoleil, A 5 × 200 Gbps microring modulator silicon chip empowered by two-segment Z-shape junctions, Nat. Commun. 15(1), 918 (2024)

[381] H. Zhou , J. Dong , J. Cheng , W. Dong , C. Huang , Y. Shen , Q. Zhang , M. Gu , C. Qian , H. Chen , Z. Ruan , and X. Zhang , Photonic matrix multiplication lights up photonic accelerator and beyond, Light Sci. Appl. 11(1), 30 (2022)

[382] X. Ouyang , Y. Xu , M. Xian , Z. Feng , L. Zhu , Y. Cao , S. Lan , B. O. Guan , C. W. Qiu , M. Gu , and X. Li , Synthetic helical dichroism for six-dimensional optical orbital angular momentum multiplexing, Nat. Photonics 15(12), 901 (2021)

[383] A. Y. Piggott , J. Lu , K. G. Lagoudakis , J. Petykiewicz , T. M. Babinec , and J. Vučković , Inverse design and demonstration of a compact and broadband on-chip wavelength demultiplexer, Nat. Photonics 9(6), 374 (2015)

[384] A. Y. Piggott , J. Petykiewicz , L. Su , and J. Vučković , Fabrication-constrained nanophotonic inverse design, Sci. Rep. 7(1), 1786 (2017)

[385] T. Lin , F. Tian , P. Shi , F. S. Chau , G. Zhou , X. Tang , and J. Deng , Design of mechanically-tunable photonic crystal split-beam nanocavity, Opt. Lett. 40(15), 3504 (2015)

[386] L. Su , D. Vercruysse , J. Skarda , N. V. Sapra , J. A. Petykiewicz , and J. Vučković , Nanophotonic inverse design with SPINS: Software architecture and practical considerations, Appl. Phys. Rev. 7(1), 011407 (2020)

[387] M. M. Islam , A. Krishnaprasad , D. Dev , R. Martinez-Martinez , V. Okonkwo , B. Wu , S. S. Han , T. S. Bae , H. S. Chung , J. Touma , Y. Jung , and T. Roy , Multiwavelength optoelectronic synapse with 2D materials for mixed-color pattern recognition, ACS Nano 16(7), 10188 (2022)

[388] F. Zhou , Z. Zhou , J. Chen , T. H. Choy , J. Wang , N. Zhang , Z. Lin , S. Yu , J. Kang , H. S. P. Wong , and Y. Chai , Optoelectronic resistive random access memory for neuromorphic vision sensors, Nat. Nanotechnol. 14(8), 776 (2019)

[389] G. Feng , J. Jiang , Y. Li , D. Xie , B. Tian , and Q. Wan , Flexible vertical photogating transistor network with an ultrashort channel for in‐sensor visual nociceptor, Adv. Funct. Mater. 31(36), 2104327 (2021)

[390] B. Cui , Z. Fan , W. Li , Y. Chen , S. Dong , Z. Tan , S. Cheng , B. Tian , R. Tao , G. Tian , D. Chen , Z. Hou , M. Qin , M. Zeng , X. Lu , G. Zhou , X. Gao , and J. M. Liu , Ferroelectric photosensor network: An advanced hardware solution to real-time machine vision, Nat. Commun. 13(1), 1707 (2022)

[391] Z. Hao , H. Wang , S. Jiang , J. Qian , X. Xu , Y. Li , M. Pei , B. Zhang , J. Guo , H. Zhao , J. Chen , Y. Tong , J. Wang , X. Wang , Y. Shi , and Y. Li , Retina‐inspired self‐powered artificial optoelectronic synapses with selective detection in organic asymmetric heterojunctions, Adv. Sci. (Weinh. ) 9(7), 2103494 (2022)

[392] Z. He , H. Shen , D. Ye , L. Xiang , W. Zhao , J. Ding , F. Zhang , C. Di , and D. Zhu , An organic transistor with light intensity-dependent active photoadaptation, Nat. Electron. 4(7), 522 (2021)

[393] C. Choi , J. Leem , M. Kim , A. Taqieddin , C. Cho , K. W. Cho , G. J. Lee , H. Seung , H. J. Bae , Y. M. Song , T. Hyeon , N. R. Aluru , S. W. Nam , and D. H. Kim , Curved neuromorphic image sensor array using a MoS₂-organic heterostructure inspired by the human visual recognition system, Nat. Commun. 11(1), 5934 (2020)

[394] Y. Gong,X. Xing,Z. Lv, ., Ultrasensitive flexible memory phototransistor with detectivity of 1.8×10¹³ Jones for artificial visual nociceptor, Adv. Intell. Syst. 4(8) (2022)

[395] X. Yang,Z. Xiong,Y. Chen,Y. Ren,L. Zhou,H. Li,Y. Zhou,F. Pan,S. T. Han, A self-powered artificial retina perception system for image preprocessing based on photovoltaic devices and memristive arrays, Nano Energy 78, 105246 (2020)

[396] S. Lee , R. Peng , C. Wu , and M. Li , Programmable black phosphorus image sensor for broadband optoelectronic edge computing, Nat. Commun. 13(1), 1485 (2022)

[397] Q. B. Zhu,B. Li,D. D. Yang,C. Liu,S. Feng,M. L. Chen,Y. Sun,Y. N. Tian,X. Su,X. M. Wang,S. Qiu,Q. W. Li,X. M. Li,H. B. Zeng,H. M. Cheng,D. M. Sun, A flexible ultrasensitive optoelectronic sensor array for neuromorphic vision systems, Nat. Commun. 12(1), 1798 (2021)

[398] M. Wang,H. Feng,D. Qi,L. Du,Z. Sha, φ-OTDR pattern recognition based on CNN-LSTM\, Optik (Stuttg.) 272, 170380 (2023)

[399] M. Sabih and D. K. Vishwakarma , Crowd anomaly detection with LSTMs using optical features and domain knowledge for improved inferring, Vis. Comput. 38(5), 1719 (2022)

[400] Q. Wang , Y. Zhang , G. Chen , Z. Chen , and H. I. Hee , Assessment of heart rate and respiratory rate for perioperative infants based on ELC model, IEEE Sens. J. 21(12), 13685 (2021)

[401] R. Xu,P. Lv,F. Xu,Y. Shi, A survey of approaches for implementing optical neural networks, Opt. Laser Technol. 136, 106787 (2021)

[402] F. Zangeneh-Nejad , D. L. Sounas , A. Alù , and R. Fleury , Analogue computing with metamaterials, Nat. Rev. Mater. 6(3), 207 (2020)

[403] T. W. Hughes , M. Minkov , Y. Shi , and S. Fan , Training of photonic neural networks through in situ backpropagation, Optica 5(7), 864 (2018)

[404] T. Hughes , G. Veronis , K. P. Wootton , R. Joel England , and S. Fan , Method for computationally efficient design of dielectric laser accelerator structures, Opt. Express 25(13), 15414 (2017)

[405] M. Miscuglio and V. J. Sorger , Photonic tensor cores for machine learning, Appl. Phys. Rev. 7(3), 031404 (2020)

[406] R. Hamerly , L. Bernstein , A. Sludds , M. Soljačić , and D. Englund , Large-scale optical neural networks based on photoelectric multiplication, Phys. Rev. X 9(2), 021032 (2019)

[407] N. Mohammadi Estakhri , B. Edwards , and N. Engheta , Inverse-designed metastructures that solve equations, Science 363(6433), 1333 (2019)

[408] N. C. Harris , J. Carolan , D. Bunandar , M. Prabhu , M. Hochberg , T. Baehr-Jones , M. L. Fanto , A. M. Smith , C. C. Tison , P. M. Alsing , and D. Englund , Linear programmable nanophotonic processors, Optica 5(12), 1623 (2018)

[409] E. Khoram , A. Chen , D. Liu , L. Ying , Q. Wang , M. Yuan , and Z. Yu , Nanophotonic media for artificial neural inference, Photon. Res. 7(8), 823 (2019)

[410] X. Sui,Q. Wu,J. Liu,Q. Chen,G. Gu, A review of optical neural networks, IEEE Access 8, 70773 (2020)

[411] D. Brunner , M. C. Soriano , C. R. Mirasso , and I. Fischer , Parallel photonic information processing at gigabyte per second data rates using transient states, Nat. Commun. 4(1), 1364 (2013)

[412] M. Hermans , M. Burm , T. Van Vaerenbergh , J. Dambre , and P. Bienstman , Trainable hardware for dynamical computing using error backpropagation through physical media, Nat. Commun. 6(1), 6729 (2015)

[413] B. J. Shastri,A. N. Tait,T. Ferreira de Lima,M. A. Nahmias,H. T. Peng,P. R. Prucnal, Principles of neuromorphic photonics, in: A. Adamatzky, (Eds.), Unconventional Computing. Encyclopedia of Complexity and Systems Science Series, Springer, New York, NY

[414] T. Yan , J. Wu , T. Zhou , H. Xie , F. Xu , J. Fan , L. Fang , X. Lin , and Q. Dai , Fourier-space diffractive deep neural network, Phys. Rev. Lett. 123(2), 023901 (2019)

[415] M. S. S. Rahman,J. Li,D. Mengu, ., Ensemble learning of diffractive optical networks, Light Sci. Appl. 10(1) (2021)

[416] Y. Luo , D. Mengu , N. T. Yardimci , Y. Rivenson , M. Veli , M. Jarrahi , and A. Ozcan , Design of task-specific optical systems using broadband diffractive neural networks, Light Sci. Appl. 8(1), 112 (2019)

[417] T. Zhou , X. Lin , J. Wu , Y. Chen , H. Xie , Y. Li , J. Fan , H. Wu , L. Fang , and Q. Dai , Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit, Nat. Photonics 15(5), 367 (2021)

[418] K. Y. Yang , J. Skarda , M. Cotrufo , A. Dutt , G. H. Ahn , M. Sawaby , D. Vercruysse , A. Arbabian , S. Fan , A. Alù , and J. Vučković , Inverse-designed non-reciprocal pulse router for chip-based LiDAR, Nat. Photonics 14(6), 369 (2020)

[419] K. Bharti , A. Cervera-Lierta , T. H. Kyaw , T. Haug , S. Alperin-Lea , A. Anand , M. Degroote , H. Heimonen , J. S. Kottmann , T. Menke , W. K. Mok , S. Sim , L. C. Kwek , and A. Aspuru-Guzik , Noisy intermediate-scale quantum (NISQ) algorithms, Rev. Mod. Phys. 94(1), 015004 (2022)

[420] C. Zhu , E. A. Bamidele , X. Shen , G. Zhu , and B. Li , Machine learning aided design and optimization of thermal metamaterials, Chem. Rev. 124(7), 4258 (2024)

[421] N. V. Sapra , K. Y. Yang , D. Vercruysse , K. J. Leedle , D. S. Black , R. J. England , L. Su , R. Trivedi , Y. Miao , O. Solgaard , R. L. Byer , and J. Vučković , On-chip integrated laser-driven particle accelerator, Science 367(6473), 79 (2020)

[422] H. Ren , W. Shao , Y. Li , F. Salim , and M. Gu , Three-dimensional vectorial holography based on machine learning inverse design, Sci. Adv. 6(16), eaaz4261 (2020)

[423] G. Qu , W. Yang , Q. Song , Y. Liu , C. W. Qiu , J. Han , D. P. Tsai , and S. Xiao , Reprogrammable meta-hologram for optical encryption, Nat. Commun. 11(1), 5484 (2020)

[424] J. P. Balthasar Mueller , N. A. Rubin , R. C. Devlin , B. Groever , and F. Capasso , Metasurface polarization optics: Independent phase control of arbitrary orthogonal states of polarization, Phys. Rev. Lett. 118(11), 113901 (2017)

[425] B. Xiong , Y. Xu , J. Wang , L. Li , L. Deng , F. Cheng , R. W. Peng , M. Wang , and Y. Liu , Realizing colorful holographic mimicry by metasurfaces, Adv. Mater. 33(21), 2005864 (2021)

[426] B. Xiong , Y. Liu , Y. Xu , L. Deng , C. W. Chen , J. N. Wang , R. Peng , Y. Lai , Y. Liu , and M. Wang , Breaking the limitation of polarization multiplexing in optical metasurfaces with engineered noise, Science 379(6629), 294 (2023)

[427] S. Wan , C. Wan , C. Dai , Z. Li , J. Tang , G. Zheng , and Z. Li , Angular-multiplexing metasurface: Building up independent-encoded amplitude/phase dictionary for angular illumination, Adv. Opt. Mater. 9(22), 2101547 (2021)

[428] W. Ma , Y. Xu , B. Xiong , L. Deng , R. W. Peng , M. Wang , and Y. Liu , Pushing the limits of functionality‐multiplexing capability in metasurface design based on statistical machine learning, Adv. Mater. 34(16), 2110022 (2022)

[429] Z. Deng , M. Jin , X. Ye , S. Wang , T. Shi , J. Deng , N. Mao , Y. Cao , B. O. Guan , A. Alù , G. Li , and X. Li , Full color complex amplitude vectorial holograms based on multi‐freedom metasurfaces, Adv. Funct. Mater. 30(21), 1910610 (2020)

[430] L. Deng , R. Jin , Y. Xu , and Y. Liu , Structured light generation using angle‐multiplexed metasurfaces, Adv. Opt. Mater. 11(16), 2300299 (2023)

[431] U. Teğin , M. Yıldırım , İ. Oğuz , C. Moser , and D. Psaltis , Scalable optical learning operator, Nat. Comput. Sci. 1(8), 542 (2021)

[432] P. R. Wiecha , A. Lecestre , N. Mallet , and G. Larrieu , Pushing the limits of optical information storage using deep learning, Nat. Nanotechnol. 14(3), 237 (2019)

[433] R. Monge , T. Delord , and C. A. Meriles , Reversible optical data storage below the diffraction limit, Nat. Nanotechnol. 19(2), 202 (2024)

[434] Y. Ma , Y. Z. Ma , Z. Q. Zhou , C. F. Li , and G. C. Guo , One-hour coherent optical storage in an atomic frequency comb memory, Nat. Commun. 12(1), 2381 (2021)

[435] M. Gu , X. Li , and Y. Cao , Optical storage arrays: A perspective for future big data storage, Light Sci. Appl. 3(5), e177 (2014)

[436] C. Wang , J. Ma , Y. Feng , X. Y. Xu , T. Y. Zhang , K. Cheng , and X. M. Jin , Error‐free long‐lifespan optical storage enhanced by deep learning, Laser Photonics Rev. 18(6), 2301042 (2024)

[437] H. Lu,B. Du,J. Liu,H. Xia,W. K. Yeap, A kernel extreme learning machine algorithm based on improved particle swam optimization, Memet. Comput. 9(2), 121 (2017)

[438] J. Zhao , Z. Wang , and F. Cao , Extreme learning machine with errors in variables, World Wide Web (Bussum) 17(5), 1205 (2014)

[439] B. He,T. Sun,T. Yan,Y. Shen,R. Nian, A pruning ensemble model of extreme learning machine with L1/2 regularizer, Multidimens. Syst. Signal Process. 28(3), 1051 (2017)

[440] J. Wang , C. Liu , and M. Zhou , Improved bacterial foraging algorithm for cell formation and product scheduling considering learning and forgetting factors in cellular manufacturing systems, IEEE Syst. J. 14(2), 3047 (2020)

[441] X. Cui , B. Wang , L. Wang , and J. Chen , Online optimal learning algorithm for Stackelberg games with partially unknown dynamics and constrained inputs, Neurocomputing 445, 1 (2021)

[442] P. Liu , Z. Liu , J. Wang , Z. Wu , P. Li , and H. Lu , Reinforcement learning empowered multi-AGV offloading scheduling in edge-cloud IIoT, J. Cloud Comput. (Heidelb. ) 11(1), 78 (2022)

[443] D. Li , Z. Luo , and B. Cao , Blockchain-based federated learning methodologies in smart environments, Cluster Computing 25(4), 2585 (2022)

[444] W. Liang , X. Chen , S. Huang , G. Xiong , K. Yan , and X. Zhou , Federal learning edge network based sentiment analysis combating global COVID-19, Comput. Commun. 204, 33 (2023)

[445] L. Lei , B. Huang , M. Ye , F. Yao , and Y. Qian , Cross-domain residual deep NMF for transfer learning between different hyperspectral image scenes, Int. J. Wavelets Multiresolution Inf. Process. 21(2), 2250046 (2023)

[446] K. Yan , X. Guo , Z. Ji , and X. Zhou , Deep transfer learning for cross-species plant disease diagnosis adapting mixed subdomains, IEEE/ACM Trans. Comput. Biol. Bioinformatics 20(4), 2555 (2023)

[447] M. Ye , C. Wang , Z. Meng , F. Xiong , and Y. Qian , Domain-invariant attention network for transfer learning between cross-scene hyperspectral images, IET Comput. Vis. 17(7), 739 (2023)

[448] X. Wang , S. Feng , and W. Yan , Human gait recognition based on self-adaptive hidden Markov model, IEEE/ACM Trans. Comput. Biol. Bioinformatics 18(3), 963 (2021)

[449] J. Sun,J. M. Garibaldi,Y. Zhang,A. Al-Shawabkeh, A multi-cycled sequential memetic computing approach for constrained optimisation, Inf. Sci. 340–341, 175 (2016)

[450] G. Xiong,K. Yan,X. Zhou, A distributed learning based sentiment analysis methods with Web applications, World Wide Web (Bussum) 25(5), 1905 (2022)

[451] F. Cao and F. Feng , Consensus‐based distributed learning for robust convex optimization with a scenario approach, Concurr. Comput. 33(8), e5049 (2021)

[452] Q. Chen and F. Cao , Distributed support vector machine in master–slave mode, Neural Netw. 101, 94 (2018)

[453] C. Wang and F. Cao , Error analysis of distributed algorithm for large scale data classification, J. Comput. Anal. Appl. 21(7), 1170 (2016)

[454] C. Liu and J. Wang , Cell formation and task scheduling considering multi-functional resource and part movement using hybrid simulated annealing, Int. J. Comput. Intelligence Systems 9(4), 765 (2016)

[455] C. Liu , J. Wang , and J. Y. T. Leung , Worker assignment and production planning with learning and forgetting in manufacturing cells by hybrid bacteria foraging algorithm, Comput. Ind. Eng. 96, 162 (2016)

[456] X. Wang , Z. Zhang , J. Li , Y. Wang , H. Cao , Z. Li , and L. Shan , An optimized encoding algorithm for systematic polar codes, EURASIP J. Wirel. Commun. Netw. 2019(1), 193 (2019)

[457] H. Yin , D. S. Park , and X. Y. Zhang , Buffer structure optimized VLSI architecture for efficient hierarchical integer pixel motion estimation implementation, J. Real-Time Image Process. 11(3), 507 (2016)

[458] J. Li , X. Wang , Z. Li , H. Wang , and L. Li , Energy efficiency optimization based on eICIC for wireless heterogeneous networks, IEEE Internet Things J. 6(6), 10166 (2019)

[459] X. Ma , Q. Jin , J. Pan , and Y. Wang , Optimization modeling and analysis of trustworthiness determination strategies for service discovery of MSNP, J. Supercomput. 75(4), 1766 (2018)

[460] G. Huang , Z. Yang , Y. Zhang , H. Peng , and J. Wang , Particle filter based optimization scheme for trajectory design and resource allocation of UAV-enabled WPCN system, Phys. Commun. 47, 101392 (2021)

[461] L. Wang , H. Wang , X. Yang , Y. Gao , X. Cui , and B. Wang , Research on smooth path planning method based on improved ant colony algorithm optimized by Floyd algorithm, Front. Neurorobot. 16, 955179 (2022)

[462] K. Huang , S. Xiu , M. Yu , X. Zhang , R. Yan , X. Yan , and Z. Liu , Software pipeline–based partitioning method with trade-off between workload balance and communication optimization, ETRI J. 37(3), 562 (2015)

[463] L. Wang , X. Li , M. Xu , Z. Guo , and B. Wang , Study on optimization model control method of light and temperature coordination of greenhouse crops with benefit priority, Comput. Electron. Agric. 210, 107892 (2023)

RIGHTS & PERMISSIONS

Higher Education Press