General Propagation Model, Architecture and System for Intelligent Optical Computing

Abstract

The present disclosure relates to the field of optical computing technologies, and in particular, to an intelligent optical computing general propagation model, architecture, and system. Among them, the model includes: a model training module, configured to obtain a target optical computing task and train an initial optical computing general propagation model according to the target optical computing task to obtain a target optical computing general propagation model, where the initial optical computing general propagation model is constructed by a four-dimensional wave optics optical field function, and the four-dimensional wave optics optical field function is used to describe the information of the complex optical field in the spatial dimension, spectral dimension, and polarization dimension; an optical computing module, configured to obtain an optical input signal and modulate the optical input signal based on the target optical computing general propagation model to obtain a modulated optical signal, so as to determine an optical computing result according to the modulated optical signal. The present disclosure adopting the above solution can improve the flexibility and practicality of optical computing.

Description

Technical Field

[0001] The present disclosure relates to the field of optical computing technologies, and in particular, to an intelligent optical computing general propagation model, architecture, and system. Background Art

[0002] With the rapid development of the fields of artificial intelligence and scientific computing, the complexity and scale of computing requirements are also increasing continuously. However, existing electronic computing technologies are limited by Moore's Law, and their performance is gradually approaching the saturation state, making it difficult to effectively meet the increasingly stringent requirements for computing power and power consumption of large-scale complex algorithms. Light has natural advantages such as high throughput and low latency during propagation. Optical computing technology using photons instead of electrons as the computing carrier is regarded as the key to breaking the existing computing bottleneck. Summary of the Invention

[0003] The present disclosure aims to solve at least one of the technical problems in the related art to some extent.

[0004] To this end, the first object of the present disclosure is to propose an intelligent optical computing general propagation model to improve the flexibility and practicality of optical computing.

[0005] The second object of the present disclosure is to propose an intelligent optical computing general propagation architecture.

[0006] The third object of the present disclosure is to propose an intelligent optical computing general propagation system.

[0007] To achieve the above object, an embodiment of the first aspect of the present disclosure proposes an intelligent optical computing general propagation model, including:

[0008] A model training module, configured to obtain a target optical computing task, and train an initial optical computing general propagation model according to the target optical computing task to obtain a target optical computing general propagation model, where the initial optical computing general propagation model is constructed by a four-dimensional wave optics light field function, and the four-dimensional wave optics light field function is used to describe the information of the complex light field in the spatial dimension, spectral dimension, and polarization dimension;

[0009] An optical computing module, configured to obtain an optical input signal, and modulate the optical input signal based on the target optical computing general propagation model to obtain a modulated optical signal, so as to determine an optical computing result according to the modulated optical signal.

[0010] Optionally, the initial optical computing general propagation model includes a first initial optical computing general propagator model in the interference dimension. Before training the initial optical computing general propagation model according to the target optical computing task, the model training module is further configured to:

[0011] Construct a first initial optical computing general propagator model for the interference dimension, where the first initial optical computing general propagator model is obtained by adding a first four-dimensional wave optics light field function and a second four-dimensional wave optics light field function.

[0012] Optionally, the initial optical computing general propagation model includes a second initial optical computing general propagator model for the diffraction dimension. Before training the initial optical computing general propagation model according to the target optical computing task, the model training module is further configured to:

[0013] Construct a second initial optical computing general propagator model for the diffraction dimension, where the second initial optical computing general propagator model is obtained by convolving a third four-dimensional wave optics light field function with a diffraction transfer function corresponding to the third four-dimensional wave optics light field function.

[0014] Optionally, the initial optical computing general propagation model includes a third initial optical computing general propagator model for the polarization dimension. Before training the initial optical computing general propagation model according to the target optical computing task, the model training module is further configured to:

[0015] Construct a third initial optical computing general propagator model for the polarization dimension, where the third initial optical computing general propagator model is obtained by performing Jones matrix multiplication on the Jones vectors of a fourth four-dimensional wave optics light field function in each wavelength channel.

[0016] Optionally, the initial optical computing general propagation model includes a fourth initial optical computing general propagator model for the spectral dimension. Before training the initial optical computing general propagation model according to the target optical computing task, the model training module is further configured to:

[0017] Construct a fourth initial optical computing general propagator model for the spectral dimension, where the fourth initial optical computing general propagator model is obtained by integrating the information of the spectral dimension and the polarization dimension of a fifth four-dimensional wave optics light field function after filtering.

[0018] Optionally, when the optical computing module is configured to modulate the optical input signal based on the target optical computing general propagation model, it is specifically configured to:

[0019] Adjust the structural parameters of the initial metasurface structure based on the target optical computing general propagation model to obtain a target metasurface structure;

[0020] Input the optical input signal into the target metasurface structure to control the target metasurface structure to modulate the optical input signal based on the target optical computing general propagation model.

[0021] Optionally, each pixel on the initial metasurface structure includes a plurality of rectangular nanocolumns. When the optical computing module is used to adjust the structural parameters of the initial metasurface structure based on the target optical computing general propagation model, it is specifically used for:

[0022] Based on the target optical computing general propagation model, determine the adjustment parameters corresponding to each rectangular nanocolumn, where the adjustment parameters include rotation angle, amplitude modulation coefficient, phase modulation coefficient, and center offset;

[0023] Based on the adjustment parameters, adjust the rectangular nanocolumns.

[0024] Optionally, when the model training module is used to train the initial optical computing general propagation model according to the target optical computing task, it is specifically used for:

[0025] Construct a loss function corresponding to the target optical computing task;

[0026] Based on the loss function, use the gradient descent method to train the model parameters in the initial optical computing general propagation model.

[0027] To achieve the above object, an embodiment of the second aspect of the present disclosure proposes an intelligent optical computing general propagation architecture, including: a data input module, the intelligent optical computing general propagation model shown in any one of the foregoing first aspects, and an output acquisition module; where

[0028] The data input module is used to obtain input data and encode the input data into the spatial dimension, spectral dimension, and polarization dimension to obtain an optical input signal;

[0029] The intelligent optical computing general propagation model is used to obtain the optical input signal and modulate the optical input signal based on the target optical computing general propagation model corresponding to the target optical computing task to obtain a modulated optical signal;

[0030] The output acquisition module is used to collect the modulated optical signal to obtain and output an optical computing result signal.

[0031] To achieve the above object, an embodiment of the third aspect of the present disclosure proposes an intelligent optical computing general propagation system, including: the intelligent optical computing general propagation architecture shown in the foregoing second aspect.

[0032] In summary, the intelligent optical computing general propagation model, architecture, and system provided by the present disclosure can provide a general optical field propagation theory and modeling method for optical computing and optical neural networks by constructing an initial optical computing general propagation model based on the four-dimensional wave optics optical field function, can improve the flexibility and practicality of optical computing, can improve the design efficiency and applicability of optical neural networks, and realize wide applications in multiple fields.

[0033] Additional aspects and advantages of the present disclosure will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above-mentioned and / or additional aspects and advantages of the present disclosure will become apparent and be readily understood from the following description of embodiments in conjunction with the accompanying drawings, in which:

[0035] Figure 1 is a schematic structural diagram of a general propagation model for intelligent optical computing provided by an embodiment of the present disclosure;

[0036] Figure 2 is a schematic structural diagram of a general propagation architecture for intelligent optical computing provided by an embodiment of the present disclosure;

[0037] Figure 3 is a schematic structural diagram of a general propagation architecture for intelligent optical computing provided by another embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] Embodiments of the present disclosure will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0039] With the rapid development of artificial intelligence, how to improve the computing speed and efficiency has become a key issue. Optical computing uses photons instead of electrons for information processing and computing, and has the advantages of high speed, low energy consumption, and parallel processing. Optical computing technology relies on physical properties such as the wavelength, polarization, diffraction, and interference of light to achieve. For example, the interference of light forms an interference pattern through the superposition of light waves, which can be used to process information. Mach-Zehnder interferometers are often used for matrix multiplication in optical computing. The diffraction characteristics of light can achieve parallel processing and operation of data by controlling the propagation path of light waves in different media. Diffractive optical neural networks can achieve tasks such as image classification at the speed of light. The polarization characteristics of light can be used for multiplexing of multiple tasks.

[0040] However, the optical propagation principles and corresponding mathematical models involved in current optical computing technologies are relatively scattered, independent of each other, lack close connection, and there is no general propagation model for intelligent optical computing. This makes it difficult to effectively combine different optical computing technologies, difficult to support the design of high-complexity optical computing systems, and difficult to meet diverse application requirements. The lack of a unified standard and general propagation model makes it difficult for different systems to be compatible, restricting the popularization and development of optical computing technology in more extensive computing applications.

[0041] The present disclosure will be described in detail below with reference to specific embodiments.

[0042] Figure 1 The following is a schematic structural diagram of a general propagation model for intelligent optical computing provided by an embodiment of the present disclosure. As Figure 1 shown, the general propagation model for intelligent optical computing includes:

[0043] A model training module, configured to obtain a target optical computing task and train an initial general propagation model for optical computing according to the target optical computing task to obtain a target general propagation model for optical computing;

[0044] An optical computing module, configured to obtain an optical input signal and modulate the optical input signal based on the target general propagation model for optical computing to obtain a modulated optical signal, so as to determine an optical computing result according to the modulated optical signal.

[0045] According to some embodiments, the target optical computing task refers to an optical computing task that needs to be performed on an optical input signal. The target optical computing task does not specifically refer to a certain fixed task.

[0046] In some embodiments, the initial general propagation model for optical computing is constructed from a four-dimensional wave optics optical field function. The four-dimensional wave optics optical field function refers to a function constructed using wave optics theory. The four-dimensional wave optics optical field function is used to describe the information of the complex optical field in the spatial dimension, spectral dimension, and polarization dimension. The four-dimensional wave optics optical field function reflects the correlation relationship between the input data and the light source characteristics.

[0047] In some embodiments, the mathematical expression of the four-dimensional wave optics optical field function can be , where , represent two spatial dimensions, represents the spectral dimension, represents the polarization dimension including , two directions.

[0048] In some embodiments, the four-dimensional wave optics optical field function can be used to construct a general propagation sub-model for optical computing in dimensions such as interference, diffraction, polarization, and spectrum.

[0049] It should be noted that the general propagation model for intelligent optical computing provided by the embodiments of the present disclosure can provide a general optical field propagation theory and modeling method for optical computing and optical neural networks by constructing an initial general propagation model for optical computing based on a four-dimensional wave optics optical field function. It can improve the flexibility and practicality of optical computing, improve the design efficiency and applicability of optical neural networks, and achieve wide applications in multiple fields.

[0050] Optionally, the initial general optical computing propagation model includes a first initial general optical computing propagator model for the interference dimension. Before training the initial general optical computing propagation model according to the target optical computing task, the model training module is further configured to:

[0051] Construct a first initial general optical computing propagator model for the interference dimension.

[0052] It should be noted that the interference of two light beams can be expressed as the direct addition corresponding to each wavelength and polarization channel of two four-dimensional light fields, that is .

[0053] That is to say, the first initial general optical computing propagator model can be obtained by adding the first four-dimensional wave optics light field function and the second four-dimensional wave optics light field function .

[0054] According to some embodiments, the modulation of the optical signal in the interference dimension can be realized by using a spatial light modulator based on the first initial general optical computing propagator model.

[0055] Optionally, the initial general optical computing propagation model includes a second initial general optical computing propagator model for the diffraction dimension. Before training the initial general optical computing propagation model according to the target optical computing task, the model training module is further configured to:

[0056] Construct a second initial general optical computing propagator model for the diffraction dimension, where the second initial general optical computing propagator model is obtained by convolving a third four-dimensional wave optics light field function with a diffraction transfer function corresponding to the third four-dimensional wave optics light field function.

[0057] For example, the diffraction of light can be expressed as the third four-dimensional light field at z = 0 , and each wavelength and polarization channel independently convolves with the diffraction transfer function d corresponding to the distance , that is:

[0058]

[0059] where represents the Rayleigh-Sommerfeld impulse response function at the j th wavelength and the k th polarization channel; i represents the imaginary symbol; r represents the distance;

[0060] It can also be converted into frequency domain calculation by using the properties of the Fourier transform, that is:

[0061]

[0062] Among them,

[0063]

[0064] According to some embodiments, by adopting a diffractive optical element, a general propagator model can be calculated based on the second initial light, and modulation of an optical signal in the interference dimension can be realized.

[0065] Optionally, the initial light calculation general propagation model includes a third initial light calculation general propagator model in the polarization dimension. Before training the initial light calculation general propagation model according to the target optical calculation task, the model training module is further configured to:

[0066] Construct a third initial light calculation general propagator model in the polarization dimension, where the third initial light calculation general propagator model is obtained by performing Jones matrix multiplication on the Jones vectors of the fourth four-dimensional wave optics light field function in each wavelength channel.

[0067] According to some embodiments, the Jones matrix multiplication can be performed according to the following formula:

[0068]

[0069] Among them, represents the polarization, intensity, and phase modulation effects of each pixel of the fourth four-dimensional wave optics light field function in the spatial dimension.

[0070] According to some embodiments, polarization, intensity, and phase modulation of an optical input signal can be realized through a metasurface micro-nano structure.

[0071] That is to say, the optical calculation module can adjust the structural parameters of the initial metasurface structure based on the target optical calculation general propagation model to obtain a target metasurface structure; input the optical input signal into the target metasurface structure to control the target metasurface structure to modulate the optical input signal based on the target optical calculation general propagation model.

[0072] In some embodiments, each pixel on the initial metasurface structure includes a plurality of rectangular nanocolumns. The optical calculation module can determine the adjustment parameters corresponding to each rectangular nanocolumn based on the target optical calculation general propagation model; and adjust the rectangular nanocolumns based on the adjustment parameters.

[0073] In some embodiments, the adjustment parameters include a rotation angle, an amplitude modulation coefficient, a phase modulation coefficient, and a center offset. The adjustment parameters can form a Jones matrix as shown in the following formula:

[0074]

[0075] Among them, represents the rotation angle The constructed rotation matrix, , and , respectively represent the amplitude modulation coefficient and phase modulation coefficient in the length and width directions of the rectangular nanorod, , represent the center offset of the rectangular nanorod.

[0076] Among them, the Jones matrix of the entire pixel containing multiple rectangular nanorods can be expressed as . The adjustment parameters of multiple rectangular nanorods are independent of each other. Therefore, 4 rectangular nanorods can realize the Jones matrix with full degrees of freedom and achieve arbitrary polarization, amplitude, and phase modulation. Similarly, the rectangular nanorods have different responses to different wavelengths, and complete spectral modulation can be achieved through special design.

[0077] Optionally, the initial light calculation general propagation model includes a fourth initial light calculation general propagator model in the spectral dimension. Before training the initial light calculation general propagation model according to the target light calculation task, the model training module is further used for:

[0078] Construct a fourth initial light calculation general propagator model in the spectral dimension.

[0079] According to some embodiments, the fourth initial light calculation general propagator model can be used to indicate the intensity of the entire light field. That is to say, this fourth initial light calculation can be obtained by integrating after filtering the information in the spectral dimension and polarization dimension of the fifth four-dimensional wave optics light field function, as shown in the following formula:

[0080]

[0081] Among them, represents the detection sensitivity of the sensor to different wavelengths and polarizations.

[0082] According to some embodiments, by using a digital micromirror device, based on the fourth initial light calculation general propagator model, modulation of the optical signal in the spectral dimension can be achieved.

[0083] Optionally, when the model training module is used to train the initial light calculation general propagation model according to the target light calculation task, it is specifically used for:

[0084] Construct a loss function corresponding to the target light calculation task;

[0085] Based on the loss function, use the gradient descent method to train the model parameters in the initial light calculation general propagation model.

[0086] It should be noted that, based on the loss function, other model training methods other than the gradient descent method can also be used to train the model parameters in the initial optical computing general propagation model.

[0087] To implement the above embodiments, the present disclosure also proposes an intelligent optical computing general propagation architecture.

[0088] As Figure 2 shown, the intelligent optical computing general propagation architecture includes: a data input module, the intelligent optical computing general propagation model provided in the foregoing embodiments, and an output acquisition module; wherein,

[0089] The data input module is used to obtain input data and encode the input data onto the spatial dimension, spectral dimension, and polarization dimension to obtain an optical input signal;

[0090] The intelligent optical computing general propagation model is used to obtain the optical input signal and modulate the optical input signal based on the target optical computing general propagation model corresponding to the target optical computing task to obtain a modulated optical signal;

[0091] The output acquisition module is used to collect the modulated optical signal to obtain and output an optical computing result signal.

[0092] According to some embodiments, the data input module can encode the input data onto the spatial dimension, spectral dimension, and polarization dimension through devices such as a spatial light modulator, a mask, a digital micromirror device, and a multi-wavelength illumination light source. Among them, encoding of the intensity dimension and phase dimension can also be realized based on the spatial dimension, spectral dimension, and polarization dimension.

[0093] In some embodiments, the output acquisition module can detect the modulated optical signal through sensors such as a photodetector and a camera.

[0094] Taking a scenario as an example, Figure 3 is a schematic structural diagram of an intelligent optical computing general propagation architecture provided by another embodiment of the present disclosure. As Figure 3 shown, it uses a multi-layer metasurface to realize the multi-dimensional optical field modulation of the intelligent optical computing general propagation model. Specifically, by combining the interference, diffraction, polarization, and spectral calculation methods involved in the intelligent optical computing general propagation model, a complete numerical propagation model from the input plane to the output plane can be constructed, and then combined with artificial intelligence algorithms to design and optimize the initial metasurface structure to construct an intelligent optical computing general propagation architecture for a specific target optical computing task.

[0095] In summary, the architecture provided in this embodiment can establish a numerical model from input to output through a general propagation model of optical computing based on dimensions such as spectrum, polarization, and space, can perform multi-dimensional modulation of the optical field, and can efficiently execute corresponding artificial intelligence tasks after training.

[0096] To implement the above embodiment, the present disclosure also proposes an intelligent optical computing general propagation system, including: the intelligent optical computing general propagation architecture provided in the foregoing embodiment.

[0097] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the present disclosure all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0098] It should be noted that personal information from users should be collected for legal and reasonable purposes and should not be shared or sold outside of these legal uses. In addition, such collection / sharing should be carried out after obtaining the informed consent of the user, including but not limited to notifying the user to read the user agreement / user notice and signing an agreement / authorization including authorizing relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others with access to the personal information data comply with their privacy policies and procedures.

[0099] The present disclosure anticipates providing embodiments that allow users to selectively block the use or access of personal information data. That is, the present disclosure anticipates providing hardware and / or software to prevent or block access to such personal information data. Once personal information data is no longer needed, the risk can be minimized by restricting data collection and deleting the data. In addition, when applicable, personal identifiers are removed from such personal information to protect the privacy of the user.

[0100] In the technical solution of the present disclosure, the acquisition, transmission, storage, use, processing, etc. of data all comply with the relevant provisions of national laws and regulations.

[0101] It should be noted that in the embodiments of the present disclosure, certain industry-existing solutions such as software, components, models, etc. may be mentioned. They should be considered exemplary, and their purpose is only to illustrate the feasibility in the implementation of the technical solution of this application, but it does not mean that the applicant has already or necessarily used this solution.

[0102] In the descriptions of the foregoing embodiments, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0103] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0104] Any process or method description in a flowchart or described in other ways herein may be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logic function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations, where the functions may be executed in a substantially simultaneous manner or in an opposite order according to the functions involved, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present disclosure pertain.

[0105] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definable list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or used in conjunction with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0106] It should be understood that various parts of the present disclosure can be implemented by hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), and the like.

[0107] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the methods of the above-described embodiments can be completed by a program instructing relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, includes one or a combination of the steps of the method embodiments.

[0108] In addition, each functional unit in various embodiments of the present disclosure may be integrated into one processing module, may exist separately as individual physical units, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0109] The above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disc, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims

1. An optical computing method based on a general propagation model of intelligent optical computing, characterized in that: include: Obtaining a target light computing task, and training an initial light computing universal propagation model according to the target light computing task to obtain a target light computing universal propagation model, wherein the initial light computing universal propagation model is constructed by a four-dimensional wave optics light field function, and the four-dimensional wave optics light field function is used to describe information of a complex light field in a spatial dimension, a spectral dimension, and a polarization dimension; Acquire an optical input signal, and modulate the optical input signal based on the target optical computing universal propagation model to obtain a modulated optical signal, so as to determine an optical computing result according to the modulated optical signal; The initial light computing universal propagation model includes a first initial light computing universal propagation sub-model in interference dimension, a second initial light computing universal propagation sub-model in diffraction dimension, a third initial light computing universal propagation sub-model in polarization dimension, and a fourth initial light computing universal propagation sub-model in spectrum dimension. Before training the initial light computing universal propagation model according to the target light computing task, the method further includes: Constructing a first initial light calculation universal propagation sub-model of interference dimension, wherein the first initial light calculation universal propagation sub-model is obtained by adding a first four-dimensional wave optics light field function and a second four-dimensional wave optics light field function; Constructing a second initial light calculation universal propagation sub-model in the diffraction dimension, wherein the second initial light calculation universal propagation sub-model is obtained by convolving a third four-dimensional wave optical light field function with a diffraction transfer function corresponding to the third four-dimensional wave optical light field function; Constructing a third initial light calculation universal propagation sub-model of polarization dimension, wherein the third initial light calculation universal propagation sub-model is obtained by performing Jones matrix multiplication operation on the Jones vector of the fourth four-dimensional wave optical light field function in each wavelength channel; Constructing a fourth initial light calculation universal propagation sub-model of spectral dimension, wherein the fourth initial light calculation universal propagation sub-model is obtained by filtering and integrating information of spectral dimension and polarization dimension of the fifth four-dimensional wave optical light field function; Among them, the mathematical expression of the four-dimensional wave optics light field function is ,in, , represents two spatial dimensions, represents the spectral dimension, Indicates that it contains , Polarization dimensions in two directions.

2. The method according to claim 1, characterized in that The step of modulating the optical input signal based on the target light calculation universal propagation model specifically includes: Adjusting the structural parameters of the initial metasurface structure based on the target light calculation universal propagation model to obtain a target metasurface structure; The optical input signal is input into the target metasurface structure to control the target metasurface structure to modulate the optical input signal based on the target light calculation universal propagation model.

3. The method according to claim 2, characterized in that Each pixel on the initial super-surface structure includes a plurality of rectangular nano-columns, and the adjusting of the structural parameters of the initial super-surface structure based on the target light calculation universal propagation model includes: Calculating a universal propagation model of the target light to determine adjustment parameters corresponding to each of the rectangular nanocolumns, wherein the adjustment parameters include a rotation angle, an amplitude modulation coefficient, a phase modulation coefficient, and a center offset; Based on the adjustment parameters, the rectangular nanorods are adjusted.

4. The method according to claim 1, characterized in that The training of the initial light computing universal propagation model according to the target light computing task includes: Constructing a loss function corresponding to the target light computing task; Based on the loss function, a gradient descent method is used to train model parameters in the initial light computation universal propagation model.

5. An optical computing device based on a general propagation model for intelligent optical computing, characterized in that: include: A data input module, a general propagation model of intelligent optical computing using the method as claimed in any one of claims 1 to 4, and an output acquisition module; wherein, The data input module is used to obtain input data and encode the input data into a spatial dimension, a spectral dimension, and a polarization dimension to obtain an optical input signal; The intelligent optical computing universal propagation model is used to obtain an optical input signal, and modulate the optical input signal based on the target optical computing universal propagation model corresponding to the target optical computing task to obtain a modulated optical signal; The output acquisition module is used to collect the modulated optical signal to obtain and output an optical calculation result signal.

6. An intelligent optical computing universal propagation system, characterized in that: include: At least one optical computing device based on the general propagation model of intelligent optical computing as claimed in claim 5.

Images