Systems, media, and methods for meta-structure grating development

Abstract

The present disclosure provides a device comprising at least one non-transitory computer readable storage medium (32) having instructions stored thereon and processing circuitry (28) coupled to the at least one non-transitory computer readable storage medium (32). The processing circuit (28) is configured to execute instructions to: provide randomized data to the neural network (400); receiving a metamaterial grating design from the neural network (400); generating a plurality of efficiency performance values based on the meta-structure grating design, each efficiency performance value included in the plurality of efficiency performance values being associated with a source angle included in a plurality of predetermined source angles; determining a loss value based on the plurality of efficiency performance values; updating the neural network based on the loss value (400); and outputting the neural network (400) to at least one of an external device or the at least one non-transitory computer readable storage medium (32).

Description

Background Technology

[0001] Electromagnetic metagratings (also known as metagratings) can modulate or otherwise influence the behavior of electromagnetic waves via deep subwavelength structures. For example, optical metagratings can modulate the behavior of wavelengths within or near the visible spectrum. Certain applications, such as augmented reality films, in-display fingerprint reader films, switchable privacy films, LiDAR and / or anti-photography films, can utilize optical metagratings. Summary of the Invention

[0002] In one embodiment, this disclosure provides an apparatus including at least one non-transitory computer-readable storage medium having instructions stored thereon and processing circuitry coupled to the at least one non-transitory computer-readable storage medium. The processing circuitry is configured to execute instructions to: provide randomized data to a neural network; receive a metagrating design from the neural network; generate a plurality of efficiency performance values ​​based on the metagrating design, including that each of the plurality of efficiency performance values ​​is associated with a source angle included in a plurality of predetermined source angles; determine a loss value based on the plurality of efficiency performance values; update the neural network based on the loss value; and output the neural network to at least one of an external device or the at least one non-transitory computer-readable storage medium.

[0003] In another embodiment, this disclosure provides an apparatus including at least one non-transitory computer-readable storage medium having instructions stored thereon and processing circuitry coupled to the at least one non-transitory computer-readable storage medium. The processing circuitry is configured to execute instructions to: provide randomized data to a neural network previously trained to generate a metagrating design by repeatedly performing the following operations: providing randomized data to the neural network; receiving the metagrating design from the neural network; generating a plurality of efficiency performance values ​​based on the metagrating design, including each of the plurality of efficiency performance values ​​being associated with a source angle included in a plurality of predetermined source angles; determining a loss value based on the plurality of efficiency performance values; and updating the neural network based on the loss value; receiving a target metagrating design from the neural network; and outputting the target metagrating design to at least one of an external device or at least one non-transitory computer-readable storage medium.

[0004] In another embodiment, this disclosure provides an apparatus including at least one non-transitory computer-readable storage medium having instructions stored thereon and processing circuitry coupled to the at least one non-transitory computer-readable storage medium. The processing circuitry is configured to execute instructions to: provide randomized data to a neural network; receive a meta-grating design from the neural network, the meta-grating design including a plurality of one-dimensional features; determine a maximum feature based on the plurality of features; shift the maximum feature within the meta-grating design; determine a loss value based on the meta-grating design; update the neural network based on the loss value; and output the neural network to at least one of an external device or at least one non-transitory computer-readable storage medium.

[0005] In another embodiment, this disclosure provides an apparatus including at least one non-transitory computer-readable storage medium having instructions stored thereon and processing circuitry coupled to the at least one non-transitory computer-readable storage medium. The processing circuitry is configured to execute instructions to: provide randomized data to a neural network previously trained to generate a metagrating design by repeatedly performing the following operations: providing randomized data to the neural network; receiving the metagrating design from the neural network; shifting the maximum feature within the metagrating design; determining a loss value based on the metagrating design; and updating the neural network based on the loss value; receiving a target metagrating design from the neural network; and outputting the target metagrating design to at least one of an external device or at least one non-transitory computer-readable storage medium.

[0006] These and additional features provided by the embodiments described herein will be more fully understood in conjunction with the accompanying drawings and the following detailed description. Attached Figure Description

[0007] Figure 1 This is a block diagram illustrating an exemplary system for utilizing and managing a communication-capable device, as exemplified by various aspects of this disclosure.

[0008] Figure 2 This is an example Figure 1 The diagram shows the system's operational perspective.

[0009] Figure 3 Exemplary metagratings according to various aspects of this disclosure are illustrated.

[0010] Figure 4 Exemplary generated networks are illustrated according to various aspects of this disclosure.

[0011] Figure 5A Exemplary meta-grating designs with discontinuous features according to various aspects of this disclosure are illustrated.

[0012] Figure 5B Examples Figure 5AThe stitching of replicas of the super-structured grating design.

[0013] Figure 5C Examples of various aspects of this disclosure include Figure 5A An exemplary meta-grating design that aggregates discontinuous features into continuous features.

[0014] Figure 6 Exemplary processes for designing shift metamorphic gratings according to various aspects of this disclosure are illustrated.

[0015] Figure 7A An exemplary diagram illustrating incident light at an angle θ and corresponding reflections using an optical meta-grating according to various aspects of this disclosure is shown.

[0016] Figure 7B An exemplary diagram illustrating incident light at an angle θ and corresponding transmission using an optical meta-grating according to various aspects of this disclosure is shown.

[0017] Figure 8 An exemplary process for training a meta-grating design generator using diffraction performance values ​​is illustrated according to various aspects of this disclosure.

[0018] Figure 9 An exemplary process for training a meta-grating design generator using efficiency performance values ​​is illustrated according to various aspects of this disclosure.

[0019] Figure 10 An exemplary process for training a meta-grating design generator using diffraction performance values ​​and / or efficiency performance values, according to various aspects of this disclosure, is illustrated.

[0020] Figure 11 Exemplary processes for generating meta-grating designs according to various aspects of this disclosure are illustrated. Detailed Implementation

[0021] Figure 1 This is a block diagram illustrating an exemplary system 2 that utilizes and manages a communication-enabled device, according to various aspects of this disclosure. System 2 includes a metagrating design system (MDS) 6 configured to provide metagrating design capabilities to a computing device 25 according to various aspects of this disclosure. As described herein, MDS 6 enables authorized users (e.g., one of users 24A to 24N) to generate metagrating designs. By interacting with MDS 6, design professionals can, for example, generate metagrating designs, train metagrating generators, and / or simulate metagrating designs. Generally, MDS 6 provides design and simulation capabilities.

[0022] like Figure 1As shown in the example, System 2 represents a computing environment in which computing devices (e.g., one of computing devices 25) can electronically communicate with MDS 6 via one or more computer networks 4. Network 4 may include one or more wired and / or wireless connections. For example, Network 4 may include connections defined by the Institute of Electrical and Electronics Engineers (IEEE) 802.11 series of protocols, ZigBee, etc. ® Network connectivity (compliant with IEEE 802.15 series standards), 5G ® Network connectivity, short-range wireless (e.g., Bluetooth) ® and / or Near Field Communication (NFC) connection, Ethernet ® Connections and / or coaxial connections.

[0023] One or more of users 24A to 24N may use computing device 25 to interact with MDS 6 via network 4. For example, end-user computing device 25 may include a laptop, desktop computer, mobile device such as a tablet or so-called smartphone, or may be part of the laptop, desktop computer, mobile device, or may be a component of the laptop, desktop computer, or mobile device.

[0024] User 24 (e.g., 24A-24N) interacts with MDS 6 to generate metagrating designs, train metagrating generators and / or models, simulate metagrating designs and / or utilize applications related to metagrating designs. For example, user 24 can generate metagrating designs to meet one or more design parameters. Furthermore, user 24 can interact with MDS 6 to simulate metagrating designs to measure the performance of one or more metagrating designs. MDS 6 enables user 24 to train generators and / or models to create metagrating designs. In some examples, MDS 6 may present a web-based interface via a web server (e.g., an HTTP server), or deploy client applications on devices such as desktop computers, laptop computers, and mobile devices such as smartphones or tablets used by user 24.

[0025] Figure 2 This is an example Figure 1 The diagram shown is an operational perspective block diagram of an example implementation of MDS 6. Although Figure 2 A specific implementation of MDS 6 consistent with various aspects of this disclosure is shown, but it will be understood that other architectures of MDS 6 (whether single-device or distributed) are also consistent with various aspects of this disclosure.

[0026] exist Figure 2In the example, MDS 6 includes one or more processors 28 and memory 32. In some examples, memory 32 and processors 28 may be integrated into a single hardware unit such as a system-on-a-chip (SoC) or integrated circuit (IC). Each processor in processor 28 may include one or more of a multi-core processor, controller, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), processing circuitry (e.g., fixed-function circuitry, programmable circuitry, or any combination of fixed-function circuitry and programmable circuitry), or equivalent discrete or integrated logic circuitry. Memory 32 may include any form of memory for storing data and executable software instructions, such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), and flash memory.

[0027] Memory 32 and processor 28 provide a computer platform for executing operating system 36. Furthermore, operating system 36 provides a multitasking operating environment for executing one or more software components 68. As shown, processor 28 is connected to external systems and devices via input / output (I / O) interface 34, such as interfaces deployed at computing device 60. I / O interface 34 may incorporate network interface hardware, such as one or more wired and / or wireless network interface controllers (NICs) for communication via communication channel 75, which may represent one or more network-enabled communication connections, such as one or more packet-switched networks. Figure 2 In the specific implementation shown, bus 70 provides inter-component connectivity between processor 28, memory 32, and I / O interface 34. Bus 70 may represent a half-duplex or full-duplex bus that provides data transfer capability between two or more of the processor 28, memory 32, I / O interface 34, and / or any other hardware component of MDS 6. Bus 70 may represent various types of system buses or computer buses, including one or more bus networks. Regardless of the implementation topology, in various examples, bus 70 may incorporate various types of inter-component connectivity hardware, such as hardware conforming to first-generation, second-generation, third-generation, or fourth-generation bus or bus network technologies as described by IEEE, and / or other bus or bus network technologies defined in standards under development or to be adopted later.

[0028] exist Figure 2In a specific example, software component 68 of MDS 6 includes a metamorphic raster design generator application 68A, a generator training application 68B, and a simulator application 68C. In some exemplary approaches, one or more software components of software component 68 represent executable software instructions, which may take the form of one or more software applications, software packages, software libraries, hardware drivers, and / or application programming interfaces (APIs). Furthermore, any software component of software component 68 may output and / or receive data via I / O interface 34.

[0029] The memory 32 provides non-volatile and / or long-term storage to support the local storage of the data repository 72. Figure 2 In the example, data repository 72 includes meta-raster designs 74A, performance metrics 74B, and analog data 74C. One or more of the software components 68 can invoke processor 28 and memory 32 to access one or more of the data repositories 72 to retrieve data for various purposes, such as comparison, processing, and relaying, and / or viewing the meta-raster designs, performance metrics, and / or analog data. In some examples, software component 68 may implement read / write capabilities regarding data repository 72, such as accessing and using information available from data repository 72 and / or modifying information currently stored in data repository 72. In a specific implementation of the distributed computing system represented by MDS 6, one or more data repositories in data repository 72 may be located at a remote location of processor 28, and in these implementations, software component 68 may use NIC hardware of I / O interface 34 to access data repository 72.

[0030] Metagrating design generator application 68A functions as an application for generating metagrating designs using a generator and / or model (e.g., a machine learning model). In some examples, metagrating design generator application 68A may also implement other processes related to metagrating generation, such as feature shifting processes. As described below, metagrating design generator application 68A can generate metagrating designs for use in various applications (e.g., optical metagratings for augmented reality films, under-display fingerprint reader films, switchable privacy films, LiDAR and / or anti-photography films). Generator training application 68B functions as an application for training machine learning models such as neural networks and / or generators to generate metagrating designs. In some examples, generator training application 68B may also implement other processes related to metagrating generation, such as feature shifting processes. In some examples, generator training application 68B may output the trained generator and / or model to metagrating design generator application 68A. The simulator application 68C is used as an application for simulating metagrating designs to generate performance metrics for a given metagrating design.

[0031] Figure 3 An exemplary metagrating 300 according to various aspects of the present disclosure is illustrated. As shown, the metagrating 300 is included in a metagrating device 304. The metagrating 300 may be disposed between a metasubstrate 308 and a substrate 312. In some examples, the metagrating 300 may be exposed to air on one or more sides, meaning that the metasubstrate 308 and / or substrate 312 may be air. In some examples, the metasubstrate 308 and / or substrate 312 may comprise one or more solid materials, such as polymers (e.g., polyethylene terephthalate and / or polyvinyl butyral) and / or silica glass. In some examples, the metasubstrate 308 and / or substrate 312 may be a uniform material having a predetermined thickness. In some examples, the metasubstrate 308 and / or substrate 312 may comprise multilayer materials, each having a predetermined thickness. In some examples, the metasubstrate 308 and / or substrate 312 may be identical. The metagrating 300 may have a predetermined thickness and may include two materials arranged in a manner that produces a desired effect (e.g., enhanced transmission efficiency at a specific wavelength and angle of incidence). Each material included in the metagrating 300 may extend throughout the entire thickness of the metagrating 300. Although the metagrating 300 is a three-dimensional material, the materials can be variably arranged along the x-axis of the metagrating 300, while the distance the materials extend along the z-axis remains constant. Therefore, the arrangement of the materials can be considered a one-dimensional problem, even though the thickness of the materials (and consequently the thickness of the metagrating 300) is a factor in the construction of the metagrating.

[0032] The physical properties of the metagrating device 304 may include: the refractive index of the material included in the metasubstrate 308 and / or substrate 312, the dispersive refractive index of the material included in the metagrating 300, the thickness of the metagrating 300, and the spacing of the metagrating 300. In some examples, if the desired thickness and / or spacing of the metagrating 300 is unknown, the thickness and / or spacing can be adjusted during training. In some examples, the mirror symmetry of the structure included in the metagrating 300 can be defined as having symmetry about the x-axis or not having symmetry. Metagrating designs utilizing symmetry can reduce training computation time by up to approximately fifty percent.

[0033] The optical characteristics of the metagrating device 304 may include the position of the light source relative to the metasubstrate 308 and / or substrate 312, optical mode (e.g., reflection, transmission, and / or absorption), polarization (e.g., transverse electric, transverse magnetic, and / or unpolarized), optical order (first or more), one or more wavelengths, one or more polar angles, one or more azimuth angles of incidence, and / or desired optical efficiency. The generator of this disclosure can be trained to generate metagratings that satisfy one or more sets of specifications defining optical characteristics. In some examples, certain optical characteristics (e.g., one or more diffraction angles) can be specified by the user, while other optical characteristics (e.g., thickness values ​​and / or spacing values) can be generated by the generator. This allows the generator to be tuned to meet specific performance requirements set by the user. In some examples, the generator can generate feature information for one or more features included in the metagrating design. In some examples, for each feature, the feature information may include material values ​​and / or position values. Material values ​​may indicate the type of material and / or material properties (e.g., wavelength values, refractive index values, and / or other optical performance information). The position value can indicate the location of the feature along the spacing of the meta-grating design.

[0034] Now for reference Figure 4 This illustration shows an exemplary generative network 400 according to various aspects of this disclosure. In some examples, the generative network 400 may be implemented as a generator and / or part of a generator. In some examples, the generative network 400 may be a neural network, such as a convolutional neural network (CNN). The generative network 400 may include multiple convolutional layers 404A-G. The generative network 400 may include multiple Leaky ReLU activations. The generative network 400 may also include periodic padding on each of the convolutional layers 404A-G. Periodic padding can incorporate periodicity into the generative network 400. The generative network 400 can generate metagratings that serve as patches in one or more periodic surfaces, so incorporating periodicity into the generative network 400 can improve the performance of the generated metagratings. In some examples, the architecture of the generative network 400 may include various numbers of layers, filter sizes, and / or scaling, and the generator training application 68B may use various network hyperparameters, such as learning rate, batch size, and / or number of steps, to train the generative network 400.

[0035] Now for reference Figure 5A , Figure 5B and Figure 5C , Figure 5A Exemplary meta-grating designs with discontinuous features according to various aspects of this disclosure are illustrated. Figure 5B Examples Figure 5A The stitching of replicas of the super-structured grating design. Figure 5C Examples of various aspects of this disclosure include Figure 5AAn exemplary metagrating design is generated by aggregating discontinuous features into continuous features. In some examples, the generator can produce metagrating designs with discontinuous features. Without considering discontinuous features, the generator might produce copies of the same metagrating design, where the only difference between metagrating designs is the periodicity of the features, thus suppressing metagrating diversity. The periodicity of the metagrating design can be taken into account by shifting discontinuous features to form continuous features.

[0036] In some examples, the training and / or generation process may include aggregating discontinuous features to form the maximum probable feature. This process may include identifying the maximum probable feature in the metagrating design and then positioning that maximum probable feature at a first end of the metagrating design. In some examples, the first end may be the left end of the metagrating design. To shift the maximum feature to the first end of the metagrating design, the process may include generating two copies of the discontinuous feature surface (e.g., Figure 5A The process involves the design of a metagrating (in the context of a supergrating design), and the stitching together of the first copy of the metagrating design, the second copy of the metagrating design, and so on. As shown in the figure... Figure 5B An example is shown of a first copy of a metagrating design stitched to a first end of the metagrating design, and a second copy of the metagrating design stitched to a second end of the metagrating design. Therefore, the process can include stitching multiple copies with discontinuous feature surfaces.

[0037] The process may include shifting the maximum feature to the first end of the metagrating design. This takes into account the periodicity of the metagrating design, and the maximum continuous feature is continuously included in the metagrating design. In some examples, the process may include detecting similar metagrating designs in a batch of metagrating designs after shifting the maximum feature using an image similarity module. After detection, relatively similar images are all shifted to have the same representation, where the maximum feature is located at the starting position of the metagrating design. It is desirable to train the generator to generate diverse shapes, rather than generating the same shapes with different periodicities. In some examples, the process may generate cosine similarity values ​​for pairs of metagrating designs, which indicate the relative similarity between the metagrating designs. The process may then evaluate the cosine similarity values ​​and / or penalize the generator based on the cosine similarity values.

[0038] Now for reference Figure 5A , Figure 5B and Figure 5C as well as Figure 6 An exemplary process 600 for shifting discontinuities according to various aspects of this disclosure is shown. In some examples, process 600 may be performed in... Figure 2The metamorphic grating design generator application 68A and / or generator training application 68B are implemented in the process. In some examples, process 600 may be implemented as at least one non-transitory computer-readable storage medium (e.g., Figure 2 Instructions are stored in memory 32 and executed by one or more processors (e.g., processor 28) coupled to the at least one non-transitory computer-readable storage medium and configured to execute the instructions.

[0039] At 604, process 600 can receive meta-grating designs (e.g., Figure 5A (Metagrating design in the context of the metagrating design). In some examples, the metagrating design may include multiple features arranged along the x-axis of the metagrating design. In some examples, the metagrating design may include feature information of one or more features included in the metagrating design. In some examples, for each feature, the feature information may include material values ​​and / or position values. Material values ​​may indicate the type of material and / or material properties (e.g., wavelength values, refractive index values, and / or other optical performance information). Position values ​​may indicate the position of the feature along the spacing of the metagrating design. In some examples, each feature may be associated with a first material or a second material. In some examples, the metagrating design may include thickness values ​​(e.g., z-axis length) and / or spacing values ​​for the width (e.g., x-axis length). Then, process 600 may proceed to 608.

[0040] At 608, process 600 can binarize the metagrating design. In some examples, at least a portion of the features included in the metagrating design can be associated with both the first material and the second material. For example, if the first material is represented by a value of zero and the second material by a value of one, the features can include values ​​selected from a continuous range from zero to one. Process 600 can binarize each feature to zero or one based on a predetermined threshold. Then, process 600 can proceed to 612.

[0041] At 612, process 600 can stitch together a first copy and a second copy of the metagrating design to the metagrating design. In some examples, process 600 can stitch together the first copy of the metagrating design to the first end of the metagrating design and stitch together the second copy of the metagrating design to the second end of the metagrating design. Then, process 600 can proceed to 616.

[0042] At 616, process 600 can determine the largest continuous feature included in the first copy of the metagrating design, the second copy of the metagrating design, and the stitching of the metagrating design. Then, process 600 can proceed to 620.

[0043] At 620, process 600 can shift the largest continuous feature to the first end of the metagrating design. In some examples, process 600 can shift the first end of the largest continuous feature to the first end of the metagrating design, thereby ensuring that the entire continuous feature is included in the metagrating design. Then, process 600 can proceed to 624.

[0044] At point 624, process 600 may output the meta-grating design to at least one of a user interface, an external device, or at least one non-transitory computer-readable storage medium. Process 600 may then terminate.

[0045] Figure 7A An exemplary diagram illustrating incident light at an angle θ and corresponding reflections using an optical meta-grating according to various aspects of this disclosure is shown. Figure 7B An exemplary diagram illustrating incident light at an angle θ and corresponding transmission using an optical meta-grating according to various aspects of this disclosure is shown.

[0046] Figure 7A and Figure 7B Several orders of refracted and transmitted light are illustrated separately. Optical films such as metagratings exhibit diffraction properties and require spectral responses suitable for specific applications. In the case of windshield composite films, it is desirable to minimize any artifacts that degrade the visual quality of the image. Light reflected or transmitted at an angle equal to the angle of incidence (assuming the metasubstrate and substrate layers are approximately identical) is referred to as specular or zero-order diffraction. Light diffracted at other angles is described as diffuse or non-zero diffraction order, and... Figure 7A and Figure 7B The values ​​are labeled + / -1 and 2. Non-zero diffraction order not only prevents preferred specular reflection and transmission at specific wavelengths and angles, but also causes perceived artifacts in optical films similar to haze and rainbow effects. Haze is the percentage of transmitted light penetrating the sample, which deviates from the incident light by no more than 0.044 radians through forward scattering. The rainbow effect refers to the phenomenon where metagratings produce different responses at different wavelengths due to material dispersion, typically measured by non-zero order reflection and transmission. As described below, the non-zero order diffraction efficiency can be estimated and used to generate loss values ​​to train a generator to produce metagrating designs that minimize the efficiency of the non-zero diffraction order. In some examples, the non-zero order diffraction efficiency can be estimated and used to generate loss values ​​to train a generator to produce metagrating designs that maximize the efficiency of the non-zero diffraction order.

[0047] Figure 8 An exemplary process 800 for training a metagrating design generator according to various aspects of this disclosure is illustrated. Specifically, process 800 may train the generator to generate metagrating designs with improved spectral responses. In some examples, the metagrating design may include a metagrating (e.g., Figure 3Metagrating 300 in the example. In some examples, the metagrating design may include a metagrating device and / or a portion thereof (e.g., Figure 3 The metagrating device 304 in the image is associated with and / or a portion thereof. In some examples, the metagrating design may include a metasubstrate (e.g., Figure 3 Super substrate 308) and substrate (e.g., Figure 3 The substrate 312 in the image. In some examples, the metagrating device and / or portions of the metagrating device may be predetermined. In some examples, the metagrating may be exposed to air on one or more sides, meaning the metasubstrate and / or substrate may be air. In some examples, the metasubstrate and / or substrate may comprise one or more solid materials, such as polymers (e.g., polyethylene terephthalate and / or polyvinyl butyral) and / or silica glass. In some examples, the metasubstrate and / or substrate may be a uniform material with a predetermined thickness. In some examples, the metasubstrate and / or substrate may comprise multilayer materials, each with a predetermined thickness. In some examples, the metasubstrate and / or substrate may be identical. The metagrating may have a predetermined thickness and may comprise two materials arranged to produce a desired effect (e.g., enhanced transmission efficiency at a specific wavelength and angle of incidence). In some examples, a generator may be trained to output a metagrating design for a predetermined metagrating device (e.g., a predetermined substrate and metasubstrate). In some examples, process 800 may be... Figure 2 The generator training is implemented in the 68B application and / or the simulator application in the 68C application.

[0048] In some examples, process 800 may be implemented as at least one non-transitory computer-readable storage medium (e.g., Figure 2 Instructions are stored in memory 32 and executed by one or more processors (e.g., processor 28) coupled to the at least one non-transitory computer-readable storage medium and configured to execute the instructions. In some examples, process 800 may be executed to train a generator. In some examples, the generator may include a machine learning model, such as a neural network (e.g., ...). Figure 4 Generative Network 400 (in the context of the network).

[0049] At 804, process 800 can receive one or more non-zero diffraction orders. In some examples, one or more non-zero diffraction orders can be received from the user at the user interface. In some examples, the user can select one or more non-zero diffraction orders as orders of interest for minimization or maximization.

[0050] In some examples, process 800 may receive one or more physical parameter values. In some examples, these one or more physical parameter values ​​may be referred to as one or more physical parameter values. These one or more physical parameter values ​​may include one or more values ​​and / or ranges that the generated meta-grating design may need to follow (e.g., pre-determined application specifications). In some examples, these one or more physical parameter values ​​may include thickness values ​​(e.g., z-axis length) and / or width spacing values ​​(e.g., x-axis length). In some examples, these one or more physical parameter values ​​may include a range of values ​​for each of the x-axis spacing and / or thickness. Process 800 may then proceed to 808.

[0051] At 808, process 800 may provide randomized data to the generator. In some examples, the randomized data may be randomized noise. In some examples, the randomized data may be a string of randomized values ​​formed as a two-dimensional input matrix. In some examples, process 800 may receive a set of randomized data (e.g., a set of one hundred or more noise matrices) which can be used to train the generator. In some examples, process 800 may also provide physical parameter values ​​to the generator. Then, process 800 may proceed to 812.

[0052] At 812, process 800 can receive the metagrating design from the generator. In some examples, the metagrating design can be... Figure 3 The metagrating design 300 in the process describes a metagrating design. The metagrating design may include manufacturing data that allows the fabrication of the metagrating based on the design. Therefore, the metagrating design can be used as a blueprint for the metagrating. In some examples, the metagrating design may include x-axis spacing values, thickness values, material information, and mappings of one or more features. The mapping of one or more features may include positional data (e.g., x-coordinates) of one or more features in the metagrating. Each feature may be a continuous portion of a given material. For example, a metagrating comprising two materials may have ten features, four formed of the first material and six formed of the second material. In some examples, the metagrating design may include a grating surface representation of the metagrating. In some examples, process 800 may shift any discontinuous features in the metagrating design to create at least one larger feature. In some examples, process 800 may perform... Figure 6 At least a portion of process 600. In some examples, the generator may include a meta-grating design shifting module that performs... Figure 6 At least a portion of process 600 is performed, and the features can be automatically shifted before the output meta-grating design. Then, process 800 can proceed to 816.

[0053] At 816, process 800 can generate diffraction performance values ​​based on the meta-grating design. In some examples, each diffraction performance value can be associated with wavelength and / or source angle. In some examples, the diffraction performance values ​​can include transmission and / or reflection values. In some examples, each of the transmission and / or reflection values ​​can be a diffraction efficiency value associated with a polarization value. In some examples, process 800 can provide the meta-grating design to a simulator. In some examples, the simulator can be a physics-based simulator. In some examples, process 800 can provide additional data to the simulator. In some examples, process 800 can provide physical parameter values ​​to the simulator. In some examples, the physical parameter values ​​can include spacing and / or thickness values. In some examples, process 800 can scale the physical parameter values ​​before providing them to the simulator. In some examples, process 800 can provide a non-zero diffraction order to the simulator. In some examples, the simulator can simulate desired optical properties (e.g., polarization and / or modes).

[0054] In some examples, the simulator can generate one or more performance metrics, such as reflection and / or transmission values ​​for one or more wavelengths, angles, and / or orders. In some examples, the simulator may include RCWA simulators such as RETICOLO or S4, and / or FDTD simulators such as Lumerical or Meep. In some examples, the simulator can generate device gradients based on the metagrating design. In some examples, the device gradient may be correlated with the refractive index values ​​of the surface of the metagrating design. Process 800 can then proceed to 820.

[0055] At 820, process 800 can determine the loss value based on the diffraction performance values. In some examples, process 800 can determine the loss value based on the sum of the diffraction performance values. In some examples, process 800 can determine the loss value based on a loss function such as a Gaussian loss function and / or a softplus loss function. Process 800 can use a loss function to determine the loss value based on the sum of the diffraction performance values. Then, process 800 can proceed to 824.

[0056] At point 824, process 800 can update the generator based on the loss value. In some examples, if conditions are not yet met (e.g., a predetermined number of training epochs have not been performed, a predetermined performance value has not been achieved, etc.), process 800 can proceed to 808 to continue training the generator. Otherwise, process 800 can proceed to 828.

[0057] At point 828, process 800 may output the generator to at least one of an external device and / or at least one non-transitory computer-readable storage medium. Then, process 800 may terminate.

[0058] Now for reference Figure 9 The diagram illustrates a process 900 for training a metagrating design generator to generate metagrating designs with smaller color shifts. Color shift is a phenomenon of perceived color change in metagrating films and can be measured in the L*a*b* color space based on viewing angle. In electromagnetic simulation software relying on the RCWA method, it is impossible to simulate color shift caused by changes in viewing angle. Therefore, color shift simulation data can be generated by simulating changes in the source angle. The presence or absence of color shift can be predicted from a heatmap showing diffraction efficiency based on the source angle and incident wavelength. A heatmap of a device with high color shift contains an orientation gradient that shows the change in diffraction efficiency when the source angle changes while the wavelength remains constant. Therefore, to estimate color shift, the change in diffraction efficiency with changes in the source angle can be used.

[0059] To penalize color shift, an item can be added to the loss function. After performing electromagnetic simulations on the generated metagrating design, the variance of the transmission and reflection efficiencies for a given wavelength (randomly sampled within a specified range of wavelengths of interest) can be calculated over a specified source angle range. This item is then added to the loss function to train the generator to create metagrating designs with smaller color shifts, while simultaneously improving the defined transmission and reflection specifications, thereby enhancing the overall performance of the metagrating device.

[0060] Figure 9 An exemplary process 900 for training a metagrating design generator according to various aspects of this disclosure is illustrated. Specifically, process 900 may train the generator to generate metagrating designs with minimal color shift. In some examples, the metagrating design may include a metagrating (e.g., Figure 3 Metagrating 300 in the example. In some examples, the metagrating design may include a metagrating device and / or a portion thereof (e.g., Figure 3 The metagrating device 304 in the image is associated with and / or a portion thereof. In some examples, the metagrating design may include a metasubstrate (e.g., Figure 3 Super substrate 308) and substrate (e.g., Figure 3The substrate 312 in the image. In some examples, the metagrating device and / or portions of the metagrating device may be predetermined. In some examples, the metagrating may be exposed to air on one or more sides, meaning the metasubstrate and / or substrate may be air. In some examples, the metasubstrate and / or substrate may comprise one or more solid materials, such as polymers (e.g., polyethylene terephthalate and / or polyvinyl butyral) and / or silica glass. In some examples, the metasubstrate and / or substrate may be a uniform material with a predetermined thickness. In some examples, the metasubstrate and / or substrate may comprise multilayer materials, each with a predetermined thickness. In some examples, the metasubstrate and / or substrate may be identical. The metagrating may have a predetermined thickness and may comprise two materials arranged to produce a desired effect (e.g., enhanced transmission efficiency at a specific wavelength and angle of incidence). In some examples, a generator may be trained to output a metagrating design for a predetermined metagrating device (e.g., a predetermined substrate and metasubstrate). In some examples, process 900 may be... Figure 2 The generator training is implemented in the 68B application and / or the simulator application in the 68C application.

[0061] In some examples, process 900 may be implemented as at least one non-transitory computer-readable storage medium (e.g., Figure 2 Instructions are stored in memory 32 and executed by one or more processors (e.g., processor 28) coupled to the at least one non-transitory computer-readable storage medium and configured to execute the instructions. In some examples, process 900 may be executed to train a generator. In some examples, the generator may include a machine learning model, such as a neural network (e.g., ...). Figure 4 Generative Network 400 (in the context of the network).

[0062] At position 904, process 900 can receive multiple source angles and multiple wavelengths. In some examples, multiple source angles and multiple wavelengths can be received from the user at the user interface. In some examples, the user can select multiple source angles and multiple wavelengths as orders of interest for minimization or maximization.

[0063] In some examples, process 900 may receive one or more physical parameter values. In some examples, these one or more physical parameter values ​​may be referred to as one or more physical parameter values. These one or more physical parameter values ​​may include one or more values ​​and / or ranges that the generated meta-grating design may need to follow (e.g., specifications of a pre-determined application). In some examples, these one or more physical parameter values ​​may include thickness values ​​(e.g., z-axis length) and / or width spacing values ​​(e.g., x-axis length). In some examples, these one or more physical parameter values ​​may include a range of values ​​for each of the x-axis spacing and / or thickness. Process 900 may then proceed to 908.

[0064] At 908, process 900 can provide randomized data to the generator. In some examples, the randomized data can be randomized noise. In some examples, the randomized data can be a string of randomized values ​​formed as a two-dimensional input matrix. In some examples, process 900 can receive a set of randomized data (e.g., a set of one hundred or more noise matrices) which can be used to train the generator. In some examples, process 900 can also provide physical parameter values ​​to the generator. Then, process 900 can proceed to 912.

[0065] At position 912, process 900 can receive the metagrating design from the generator. In some examples, the metagrating design can be... Figure 3 Metagrating design 300. The metagrating design may include manufacturing data that allows the fabrication of the metagrating based on the metagrating design. Therefore, the metagrating design can be used as a blueprint for the metagrating. In some examples, the metagrating design may include x-axis spacing values, thickness values, material information, and mappings of one or more features. The mapping of one or more features may include positional data (e.g., x-coordinates) of one or more features in the metagrating. Each feature may be a continuous portion of a given material. For example, a metagrating comprising two materials may have ten features, four formed of a first material and six formed of a second material. In some examples, the metagrating design may include feature information for each feature included in the metagrating design. In some examples, for each feature, the feature information may include material values ​​and / or positional values. Material values ​​may indicate the type and / or properties of the material (e.g., wavelength values, refractive index values, and / or other optical performance information). Positional values ​​may indicate the location of the feature along the spacing of the metagrating design. In some examples, each feature may be associated with a first material or a second material. In some examples, the metagrating design may include a grating surface representation of the metagrating. In some examples, process 900 may shift any discontinuous features in the metagrating design to create at least one larger feature. In some examples, process 900 may perform... Figure 6At least a portion of process 600. In some examples, the generator may include a meta-grating design shifting module that performs... Figure 6 At least a portion of process 600 is performed, and features can be automatically shifted before the output meta-grating design. Then, process 900 can proceed to 916.

[0066] At 916, process 900 can generate efficiency performance values ​​based on the metagrating design. In some examples, each efficiency performance value can be associated with a wavelength and / or source angle included in multiple source angles and multiple wavelengths. In some examples, the efficiency performance values ​​can include transmission and / or reflection values. In some examples, the efficiency performance values ​​can be diffraction efficiency values. In some examples, each source angle included in multiple source angles can be associated with transmission and reflection values. In some examples, process 900 can provide the metagrating design to a simulator. In some examples, the simulator can be a physics-based simulator. In some examples, process 900 can provide additional data to the simulator. In some examples, process 900 can provide physical parameter values ​​to the simulator. In some examples, the physical parameter values ​​can include spacing and / or thickness values. In some examples, process 900 can scale the physical parameter values ​​before providing them to the simulator. In some examples, the simulator can simulate desired optical properties (e.g., polarization and / or modes).

[0067] In some examples, the simulator can generate one or more performance metrics, such as reflection and / or transmission values ​​for one or more wavelengths, angles, and / or orders. In some examples, the simulator may include RCWA simulators such as RETICOLO or S4, and / or FDTD simulators such as Lumerical or Meep. In some examples, the simulator can generate device gradients based on the metagrating design. In some examples, the device gradient may be correlated with the refractive index values ​​of the surface of the metagrating design. Process 900 can then proceed to 920.

[0068] At 920, process 900 can determine the loss value based on the efficiency performance value. In some examples, process 900 can determine the loss value based on the sum of the efficiency performance values. In some examples, process 900 can determine the loss value based on a loss function such as the softplus loss function. Process 900 can use a loss function to determine the loss value based on the sum of the efficiency performance values. Then, process 900 can proceed to 924.

[0069] At point 924, process 900 can update the generator based on the loss value. In some examples, if conditions are not yet met (e.g., a predetermined number of training epochs have not been performed, a predetermined performance value has not been achieved, etc.), process 900 can proceed to 908 to continue training the generator. Otherwise, process 900 can proceed to 928.

[0070] At point 928, process 900 may output the generator to at least one of an external device and / or at least one non-transitory computer-readable storage medium. Then, process 900 may terminate.

[0071] Figure 10 An exemplary process 1000 for training a metagrating design generator according to various aspects of this disclosure is illustrated. Specifically, process 1000 may train the generator to generate metagrating designs with improved spectral response and minimal color shift. In some examples, the metagrating design may include a metagrating (e.g., Figure 3 Metagrating 300 in the example. In some examples, the metagrating design may include a metagrating device and / or a portion thereof (e.g., Figure 3 The metagrating device 304 in the image is associated with and / or a portion thereof. In some examples, the metagrating design may include a metasubstrate (e.g., Figure 3 Super substrate 308) and substrate (e.g., Figure 3 The substrate 312 in the example. In some examples, the metagrating device and / or portions of the metagrating device may be predetermined. In some examples, the metagrating may be exposed to air on one or more sides, meaning the metasubstrate and / or substrate may be air. In some examples, the metasubstrate and / or substrate may comprise one or more solid materials, such as polymers (e.g., polyethylene terephthalate and / or polyvinyl butyral) and / or silica glass. In some examples, the metasubstrate and / or substrate may be a uniform material with a predetermined thickness. In some examples, the metasubstrate and / or substrate may comprise multilayer materials, each with a predetermined thickness. In some examples, the metasubstrate and / or substrate may be identical. The metagrating may have a predetermined thickness and may comprise two materials arranged to produce a desired effect (e.g., enhanced transmission efficiency at a specific wavelength and angle of incidence). In some examples, a generator may be trained to output a metagrating design for a predetermined metagrating device (e.g., a predetermined substrate and metasubstrate). In some examples, process 1000 may be... Figure 2 The generator training is implemented in the 68B application and / or the simulator application in the 68C application.

[0072] In some examples, process 1000 may be implemented as at least one non-transitory computer-readable storage medium (e.g., Figure 2 Instructions are stored in memory 32 and executed by one or more processors (e.g., processor 28) coupled to the at least one non-transitory computer-readable storage medium and configured to execute the instructions. In some examples, process 1000 may be executed to train a generator. In some examples, the generator may include a machine learning model, such as a neural network (e.g., ...). Figure 4 Generative Network 400 in the middle.

[0073] At position 1004, process 1000 can receive multiple metagrating application parameter values. In some examples, the multiple metagrating application parameter values ​​may include, for example, in... Figure 8 One or more non-zero diffraction orders received at position 804 and in Figure 9 Multiple source angles and multiple wavelengths are received at position 904. In some examples, process 1000 may receive one or more physical parameter values, such as... Figure 8 804 and / or Figure 9 As described in 904. The one or more physical parameter values ​​may be included in multiple meta-grating application parameter values. Then, process 1000 can proceed to 1008.

[0074] At point 1008, process 1000 can provide randomized data to the generator. In some examples, this randomized data can be, for example... Figure 8 808 and / or Figure 9 The randomized noise described in 908. In some examples, process 1000 may also provide physical parameter values ​​to the generator. Then, process 1000 may proceed to 1012.

[0075] At position 1012, process 1000 can receive the metagrating design from the generator. In some examples, the metagrating design can be... Figure 3Metagrating design 300. The metagrating design may include manufacturing data that allows the fabrication of the metagrating based on the metagrating design. Therefore, the metagrating design can be used as a blueprint for the metagrating. In some examples, the metagrating design may include x-axis spacing values, thickness values, material information, and mappings of one or more features. The mapping of one or more features may include positional data (e.g., x-coordinates) of one or more features in the metagrating. Each feature may be a continuous portion of a given material. For example, a metagrating comprising two materials may have ten features, four formed of a first material and six formed of a second material. In some examples, the metagrating design may include feature information for each feature included in the metagrating design. In some examples, for each feature, the feature information may include material values ​​and / or positional values. Material values ​​may indicate the type and / or properties of the material (e.g., wavelength values, refractive index values, and / or other optical performance information). Positional values ​​may indicate the location of the feature along the spacing of the metagrating design. In some examples, each feature may be associated with a first material or a second material. In some examples, the metagrating design may include a grating surface representation of the metagrating. In some examples, process 1000 may shift any discontinuous features in the metagrating design to create at least one larger feature. In some examples, process 1000 may perform... Figure 6 At least a portion of process 600. In some examples, the generator may include a meta-grating design shifting module that performs... Figure 6 At least a portion of process 600 is performed, and features can be automatically shifted before the output meta-grating design. Then, process 1000 can proceed to 1016.

[0076] At position 1016, process 1000 can generate performance values ​​based on the metagrating design. In some examples, the performance values ​​may include, for example... Figure 8 The diffraction performance values ​​described in 816 and Figure 9 The efficiency performance value is 916. Then, process 1000 can proceed to 1020.

[0077] At point 1020, process 1000 can determine the loss value based on the performance value. In some examples, this loss value can be based on, for example... Figure 8 The loss value determined by the efficiency performance value and diffraction performance value described in section 816 and based on Figure 9 The efficiency performance value 916 determines the final loss value. Then, process 1000 can proceed to 1024.

[0078] At point 1024, process 1000 can update the generator based on the loss value. In some examples, if conditions are not yet met (e.g., a predetermined number of training epochs have not been performed, a predetermined performance value has not been achieved, etc.), process 1000 can proceed to 1008 to continue training the generator. Otherwise, process 1000 can proceed to 1028.

[0079] At point 1028, process 1000 may output the generator to at least one of an external device and / or at least one non-transitory computer-readable storage medium. Then, process 1000 may terminate.

[0080] Figure 11 An exemplary process 1100 for generating a metagrating design according to various aspects of this disclosure is illustrated. Specifically, process 1100 can generate a metagrating design that can be fabricated for a variety of metagrating devices. In some examples, the metagrating design may include a metagrating (e.g., Figure 3 Metagrating 300 in the example. In some examples, the metagrating design may include a metagrating device and / or a portion thereof (e.g., Figure 3 The metagrating device 304 in the image is associated with and / or a portion thereof. In some examples, the metagrating design may include a metasubstrate (e.g., Figure 3 Super substrate 308) and substrate (e.g., Figure 3 The substrate 312 in the image. In some examples, the metagrating device and / or portions of the metagrating device may be predetermined. In some examples, the metagrating may be exposed to air on one or more sides, meaning the metasubstrate and / or substrate may be air. In some examples, the metasubstrate and / or substrate may comprise one or more solid materials, such as polymers (e.g., polyethylene terephthalate and / or polyvinyl butyral) and / or silica glass. In some examples, the metasubstrate and / or substrate may be a uniform material with a predetermined thickness. In some examples, the metasubstrate and / or substrate may comprise multilayer materials, each with a predetermined thickness. In some examples, the metasubstrate and / or substrate may be identical. The metagrating may have a predetermined thickness and comprise two materials arranged to produce a desired effect (e.g., enhanced transmission efficiency at a specific wavelength and incident angle). In some examples, a generator may be trained to output a metagrating design for a predetermined metagrating device (e.g., a predetermined substrate and metasubstrate). In some examples, process 1100 may be... Figure 2 The meta-grating design generator application 68A, the generator training application 68B, and / or the simulator application 68C are implemented in the application.

[0081] In some examples, process 1100 may be implemented as at least one non-transitory computer-readable storage medium (e.g., Figure 2 Instructions are stored in memory 32 and executed by one or more processors (e.g., processor 28) coupled to the at least one non-transitory computer-readable storage medium and configured to execute the instructions.

[0082] At 1104, process 1100 may receive one or more metagrating application parameter values. In some examples, these one or more metagrating application parameter values ​​may be selected by the user. In some examples, these one or more metagrating application parameter values ​​may include one or more physical parameter values ​​and / or performance parameter values. In some examples, the one or more physical parameter values ​​may be... Figure 4 The physical parameter value is 408. In some examples, this one or more physical parameter values ​​may be referred to as one or more physical parameter values. This one or more physical parameter values ​​may include one or more values ​​and / or ranges that the resulting meta-grating design may need to follow (e.g., specifications of a pre-determined application). In some examples, this one or more physical parameter values ​​may include physical parameter values ​​such as thickness values ​​(e.g., z-axis length) and / or width spacing values ​​(e.g., x-axis length). In some examples, this one or more physical parameter values ​​may include ranges of values ​​for each of the x-axis spacing and thickness.

[0083] In some examples, performance parameter values ​​may include Figure 4 One or more of the user-defined metagrating specifications 416 in the specification can be used. Performance parameter values ​​(e.g., by the user) can be selected to train the generator to produce metagrating designs with the desired performance quality. In some examples, performance parameter values ​​may include one or more wavelength, angle, and / or order values ​​for reflection, transmission, and / or absorption. In some examples, performance parameter values ​​can be generated using Equations 2 and / or 3 described above. Process 1100 can then proceed to 1108.

[0084] At 1108, process 1100 may select a generator based on the one or more metagrating application parameter values. In some examples, process 1100 may select a trained generator that satisfies each of the one or more metagrating application parameter values. In some examples, process 1100 may select a generator from a database of pre-trained generators. In some examples, process 1100 may train the generator to generate metagrating designs that satisfy each of the one or more metagrating application parameter values. In some examples, process 1100 may perform... Figure 8 The process in 800 Figure 9 In process 900 and / or Figure 10At least a portion of process 1000 is used to train the generator by applying parameter values ​​using one or more meta-gratings. Once the generator has been selected and / or trained, process 1100 can proceed to 1112.

[0085] At 1112, process 1100 may provide randomized data to the generator. In some examples, process 1100 may receive randomized data from a user and / or a database. In some examples, the randomized data may be randomized noise. In some examples, the randomized data may be a string of random values ​​formed as a two-dimensional input matrix. Then, process 1100 may proceed to 1116.

[0086] At 1116, process 1100 can receive the metagrating design from the generator. In some examples, the metagrating design can be... Figure 4 Metagrating design 304. A metagrating design may include manufacturing data that allows the fabrication of a metagrating based on the metagrating design. Therefore, a metagrating design can be used as a blueprint for a metagrating. In some examples, a metagrating design may include x-axis spacing values, thickness values, material information, and mappings of one or more features. The mapping of one or more features may include positional data (e.g., x and y coordinates) of one or more features in the metagrating. Each feature may be a continuous portion of a given material. For example, a metagrating comprising two materials may have ten features, where four features are formed of the first material and six features are formed of the second material. In some examples, a metagrating design may include a grating surface representation of the metagrating.

[0087] In some examples, process 1100 can shift any discontinuous features in the metagrating design to create at least one larger feature. In some examples, process 1100 can perform... Figure 6 At least a portion of process 600. Then, process 1100 can proceed to 1120.

[0088] At 1120, process 1100 may output the meta-grating design to at least one of a user interface, an external device, or the at least one non-transitory computer-readable storage medium. Then, process 1100 may terminate.

[0089] In the detailed description of exemplary embodiments of the invention, reference is made to the accompanying drawings, which illustrate specific embodiments in which the invention may be practiced. These exemplary embodiments are not intended to exhaustively list all embodiments according to the invention. It should be understood that other embodiments may be utilized, and structural or logical changes may be made, without departing from the scope of the invention. Therefore, the following detailed description should not be considered limiting, and the scope of the invention is defined by the appended claims.

[0090] Unless otherwise specified, all numbers used in this specification and claims to express characteristic dimensions, quantities, and physical properties should in all cases be understood to be modified by the terms “about,” “approximately,” or “substantially.” Therefore, unless stated to the contrary, the numerical parameters listed in the foregoing specification and appended claims are approximations, which may vary depending on the desired properties sought by those skilled in the art using the teachings disclosed herein.

[0091] Unless otherwise expressly stated in the description and appended claims, the singular forms “a” and “described” as used in the specification and appended claims cover embodiments having multiple referents. Unless otherwise expressly stated in the description and appended claims, the term “or” is generally used in its meaning including “and / or”.

[0092] It should be recognized that, based on this example, certain actions or events of any of the methods described herein may be performed in a different order, and may be added together, combined, or omitted (e.g., not all described actions or events are necessary for the practice of the method). Furthermore, in some examples, actions or events may be executed simultaneously rather than sequentially, for example, through multithreaded processing, interrupt handling, or multiple processors.

[0093] The techniques described in this disclosure can be implemented, at least in part, in hardware, software, firmware, or any combination thereof. For example, aspects of the described techniques can be implemented in one or more processors, including one or more microprocessors, CPUs, GPUs, DSPs, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or any other equivalent integrated or discrete logic circuits, and any combination of such components. The terms "processor" or "processing circuitry" can generally refer to any of the aforementioned logic circuits (e.g., fixed-function circuitry, programmable circuitry, or any combination of fixed-function and programmable circuitry), alone or in combination with other logic circuitry, or any other equivalent circuitry. A control unit containing hardware can also perform one or more of the techniques disclosed herein.

[0094] Such hardware, software, and firmware may be implemented within the same device or in different devices to support the various operations and functions described in this disclosure. Furthermore, any described unit, module, or component may be implemented together or individually as discrete but cooperating logical devices. Describing different features as modules or units is intended to highlight different functional aspects and does not necessarily imply that such modules or units must be implemented by separate hardware or software components. Rather, the functionality associated with one or more modules or units may be performed by separate hardware or software components or integrated into common or separate hardware or software components.

[0095] The techniques described in this disclosure may also be embodied or encoded in a computer-readable medium, such as a computer-readable storage medium, containing instructions. For example, when the instructions are executed, instructions embedded or encoded in a computer-readable storage medium may cause a programmable processor or other processor to perform the method. Computer-readable storage media may include: random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, hard disk, CD-ROM, floppy disk, magnetic tape cassette, magnetic media, optical media, or other computer system-readable media.

[0096] Various examples have been described. These examples, as well as others, are all within the scope of the following claims.

Claims

1. An apparatus, the apparatus comprising: At least one non-transitory computer-readable storage medium, on which instructions are stored; and processing circuitry coupled to the at least one non-transitory computer-readable storage medium, the processing circuitry being configured to execute the instructions to: Randomized data is fed into the neural network; Receive metagrating design from the neural network; Based on the meta-grating design, multiple efficiency performance values ​​are generated, including each efficiency performance value being associated with a source angle included in a plurality of predetermined source angles; The loss value is determined based on the multiple efficiency performance values; The neural network is updated based on the loss value; as well as The neural network outputs to at least one of an external device or at least one of the at least one non-transitory computer-readable storage media.

2. The device of claim 1, wherein each of the plurality of efficiency performance values ​​is further associated with a wavelength included in a plurality of predetermined wavelengths.

3. The apparatus of claim 2, wherein the processing circuitry is configured to further execute the instructions to: Receive the plurality of predetermined wavelengths from the user interface.

4. The apparatus of claim 1, wherein the processing circuitry is configured to further execute the instructions to: Receive the multiple predetermined source angles from the user interface.

5. The device according to claim 1, wherein the plurality of efficiency performance values ​​includes a transmission efficiency performance value.

6. The device according to claim 1, wherein the plurality of efficiency performance values ​​includes a reflection efficiency performance value.

7. The device of claim 1, wherein the plurality of efficiency performance values ​​comprises a plurality of reflection efficiency performance values ​​and a plurality of transmission efficiency performance values, wherein each of the plurality of predetermined source angles is associated with at least one reflection efficiency performance value and at least one transmission efficiency performance value among the plurality of reflection efficiency performance values.

8. The device of claim 7, wherein each of the plurality of efficiency performance values ​​is further associated with a wavelength included in a plurality of predetermined wavelengths, and each wavelength included in the plurality of predetermined wavelengths is associated with at least one reflection efficiency performance value included in the plurality of reflection efficiency performance values ​​and at least one transmission efficiency performance value included in the plurality of transmission efficiency performance values.

9. The device according to claim 1, wherein generating the plurality of efficiency performance values ​​based on the meta-grating design includes: The meta-grating design is provided to the simulator; as well as Receive the plurality of efficiency performance values ​​from the simulator.

10. The device of claim 1, wherein determining the loss value based on the plurality of efficiency performance values ​​comprises: The variance value is calculated based on the multiple efficiency performance values.

11. The device of claim 10, wherein determining the loss value based on the plurality of efficiency performance values ​​further comprises: The loss value is calculated based on the variance using the loss formula.

12. The device according to claim 11, wherein the loss formula is a softplus formula.

13. The apparatus of claim 1, wherein the processing circuitry is configured to further execute the instructions to: Based on the meta-grating design, diffraction performance values ​​are generated for each of the multiple non-zero orders; A second loss value is determined based on the diffraction performance values ​​associated with the plurality of non-zero orders; and The neural network is further updated based on the second loss value.

14. The apparatus of claim 13, wherein generating the diffraction performance value for each of the plurality of non-zero orders comprises: The meta-grating design is provided to the simulator; as well as The simulator receives the diffraction performance value associated with each of the plurality of non-zero orders.

15. The device of claim 13, wherein determining the second loss value comprises: Calculate the sum of the diffraction performance values ​​associated with the plurality of non-zero orders; as well as The second loss value is determined using a loss function based on the sum of the diffraction performance values.

16. The device of claim 15, wherein the loss function is a Gaussian loss function.

17. The device of claim 15, wherein the loss function is a softplus loss function.

18. The device according to claim 13, wherein the diffraction performance value is a diffraction efficiency value.

19. The device of claim 13, wherein each of the diffraction performance values ​​associated with the plurality of non-zero orders is further associated with a predetermined wavelength.

20. The device of claim 13, wherein each of the diffraction performance values ​​associated with the plurality of non-zero orders is further associated with a predetermined angle.

21. The device of claim 13, wherein the metagrating design includes a spacing value, a thickness value, and feature information associated with at least one feature included in the metagrating design.

22. The device of claim 21, wherein for each feature included in the meta-grating design, the feature information includes a material value and a position value.

23. The device of claim 13, wherein the diffraction performance value includes at least one of a transmission value or a reflection value.

24. The device of claim 23, wherein each of the at least one of the transmission value or reflection value is a diffraction efficiency value associated with the polarization value.

25. The device of claim 13, wherein the meta-grating design is an optical film design.

26. An apparatus, the apparatus comprising: At least one non-transitory computer-readable storage medium, on which instructions are stored; and processing circuitry coupled to the at least one non-transitory computer-readable storage medium, the processing circuitry being configured to execute the instructions to: Randomized data is fed into a neural network that has previously been trained to generate metagrating designs by repeatedly performing the following operations: Randomized data is provided to the neural network; Receive metagrating design from the neural network; Based on the meta-grating design, multiple efficiency performance values ​​are generated, including each efficiency performance value being associated with a source angle included in a plurality of predetermined source angles; The loss value is determined based on the multiple efficiency performance values; as well as The neural network is updated based on the loss value; The target meta-grating design is received from the neural network; as well as The target meta-grating design is output to an external device or at least one of the at least one non-transitory computer-readable storage media.

27. The device of claim 26, wherein the neural network has also been previously trained to generate a meta-grating design by repeatedly performing the following operations: Based on the meta-grating design, diffraction performance values ​​are generated for each of the multiple non-zero orders; The second loss value is determined based on the diffraction performance values ​​associated with the plurality of non-zero orders; as well as The neural network is further updated based on the second loss value.

28. An apparatus, the apparatus comprising: At least one non-transitory computer-readable storage medium, on which instructions are stored; and processing circuitry coupled to at least one non-transitory computer-readable storage medium, said processing circuitry being configured to execute the instructions to: Randomized data is fed into the neural network; The neural network receives a metagrating design, the metagrating design comprising multiple one-dimensional features; The largest feature is determined based on the aforementioned multiple features; Displace the largest feature within the meta-grating design; The loss value is determined based on the aforementioned meta-grating design; The neural network is updated based on the loss value; as well as The neural network outputs to at least one of an external device or at least one of the at least one non-transitory computer-readable storage media.

29. The device of claim 28, wherein determining the most prominent feature comprises stitching together multiple copies of the meta-grating design.

30. The device of claim 28, wherein determining the most prominent feature comprises: A first copy of the meta-grating design is spliced ​​to the first end of the meta-grating design; A second copy of the metagrating design is spliced ​​to the second end of the metagrating design; and The detection includes the largest continuous one-dimensional feature in at least one of the metagrating design, the first copy of the metagrating design, or the second copy of the metagrating design.

31. The device of claim 30, wherein the most prominent feature within the shifted meta-grating design comprises: The first end of the largest continuous one-dimensional feature is shifted to the first end of the meta-grating design.

32. The apparatus of claim 28, wherein the processing circuitry is configured to further execute the instructions to: The secondary randomized data is provided to the neural network; The neural network receives a second metagrating design, the second metagrating design comprising a second plurality of one-dimensional features; The second largest feature is determined based on the second set of features; Displace the second largest feature within the second meta-grating design; as well as The loss value is further determined based on the second meta-grating design.

33. The apparatus of claim 32, wherein the processing circuitry is configured to further execute the instructions to: Calculate the cosine similarity value based on the metagrating design and the second metagrating design; and The loss value is further determined based on the second meta-grating design.

34. The apparatus of claim 28, wherein the processing circuitry is configured to further execute the instructions to: Binarization of the meta-grating design.

35. The apparatus of claim 28, wherein the processing circuitry is configured to further execute the instructions to: Based on the metagrating design, multiple efficiency performance values ​​are generated, including each efficiency performance value being associated with a source angle included in a plurality of predetermined source angles; and The loss value is determined based on the plurality of efficiency performance values.

36. An apparatus, the apparatus comprising: At least one non-transitory computer-readable storage medium, on which instructions are stored; and processing circuitry coupled to the at least one non-transitory computer-readable storage medium, the processing circuitry being configured to execute the instructions to: Randomized data is fed into a neural network that has previously been trained to generate metagrating designs by repeatedly performing the following operations: Randomized data is provided to the neural network; Receive metagrating design from the neural network; Displace the largest feature within the meta-grating design; The loss value is determined based on the aforementioned meta-grating design; as well as The neural network is updated based on the loss value. Receive the target meta-grating design from the neural network; and The target meta-grating design is output to an external device or at least one of the at least one non-transitory computer-readable storage media.

37. The device of claim 36, wherein the neural network has further been previously trained to generate a meta-grating design by repeatedly performing the following operations: Based on the metagrating design, multiple efficiency performance values ​​are generated, including each efficiency performance value being associated with a source angle included in a plurality of predetermined source angles; and The loss value is determined based on the plurality of efficiency performance values.

Images