Method and device for training modular network auto-encoder based on isovariant constraint
Through isovariable constraints, the problem of insufficient functional specialization in the existing technology is solved, and more effective and stable functional differentiation of modular networks is achieved, and the robustness and universality of downstream tasks is improved. ability to transform.
Patent Information
- Application Number
- CN202410940666.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-13
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-13
AI Technical Summary
The existing network modularization methods have bottlenecks in functional specialization, making it difficult to achieve more effective, more stable and more controllable functional specialization, especially when facing new tasks or data types, lack flexibility and scalability.
The modular network autoencoder training method based on isovariant constraints is adopted to train the convolutional network autoencoder through isovariant constraints, and the convolution kernel module is optimized by weight coefficients of isovariant loss and symmetric loss, so as to enhance the correlation within the module. and independence, thereby achieving functional differentiation.
It improves the specialization degree of the functions of the autoencoder module and the robustness and generalization of downstream tasks, provides a better representation space, and is suitable for tasks such as classification, detection, segmentation and associative generation.
Smart Images

Figure CN119026638A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a training method and device for a modular network autoencoder based on equivariant constraints. Background Art
[0002] At present, artificial neural networks have been rapidly developed and widely used. The modularization of artificial neural networks has many advantages. For example, modular design can reduce the number of network parameters, thereby reducing the storage space and computational complexity of the model; it can make the network easier to parallelize, thereby increasing the computational speed of the model; it can more effectively utilize computing resources, such as GPU memory and bandwidth; it can make the network structure clearer and easier to understand the internal working mechanism of the model; it can make different models share the same modules, thereby improving the generalization ability of the model; different modules can be combined into new models, thereby achieving more complex network structures and functions. In short, network modular design can effectively improve model efficiency, interpretability, and generalization ability, and can adapt to a variety of tasks. It is an important method for building complex, efficient, and interpretable deep learning models.
[0003] There are two main existing network modularization methods:
[0004] Modular structure and manually defined functions: This type of method achieves functional differences between different modules by manually designing different modules and training them with data of different modalities or different tasks. For example, a module trained with image data may be better at recognizing visual features, while a module trained with text data may be better at understanding semantic information. The disadvantage of this method is that it is highly dependent on manual design, which limits the scalability of network modularity. When encountering new tasks or data types, the modules need to be redesigned, which is labor-intensive and lacks flexibility.
[0005] Modular structure and automatic function learning: This type of method divides the network structure into different modules, and then allows the network to automatically learn the functional specialization of different modules during training. For example, group convolution can divide the convolution kernel into different groups, each group is responsible for extracting a specific type of feature. Expert ensemble (MoE) can combine multiple expert networks for ensemble learning. Capsule networks connect different types of capsules through a dynamic routing mechanism to achieve higher-level feature extraction. Although these methods improve network efficiency and reduce the number of parameters, it is difficult to produce stable and reliable functional specialization.
[0006] It can be seen that the existing network modularization methods still have some bottlenecks in functional specialization. Therefore, how to achieve more effective, stable and controllable functional specialization is an important direction for future modular network research. Summary of the invention
[0007] In order to solve the problems existing in the prior art, the present invention provides the following technical solutions.
[0008] A first aspect of the present invention provides a training method for a modular network autoencoder based on equivariant constraints, comprising:
[0009] Acquire training data, wherein the training data includes an image I and an image I′ obtained after the image I is transformed;
[0010] Based on the training data, the convolutional network autoencoder is trained using equivariant constraints to obtain a trained autoencoder; wherein the equivariant constraint L EC The calculation is performed using the following formula:
[0011]
[0012] L equ =∑ i ||f′ (i) -M (i) (δ)f (i) || 2 , i∈{0, 1, ..., k-1},
[0013]
[0014] L EC =L recon +λ1L equ +λ2L sym ,
[0015] f=I*W,
[0016] f′=I′*W,
[0017] Where I′=L δ (I), Lδ is a transformation with a transformation parameter δ; W is the convolution kernel of the convolutional network autoencoder, and W = [W (0) , W (1) , ..., W (k-1) ], that is, the convolution kernel is divided into k modules; f is the feature image obtained after the image I is convolved, and f = [f (0) , f (1) , ..., f (k-1) ]; f′ is the feature image obtained after convolution of image I′, f′=[f′ (0) , f′ (1) , ..., f′ (k-1) ]; * indicates convolution operation, represents the deconvolution operation; f (i) is the feature map obtained after image I is convolved by the i-th module of the convolution kernel; f′(i) is the feature map obtained after the image I′ is convolved by the i-th module of the convolution kernel; M (i) (δ) is the prediction matrix of the i-th module with parameter δ, which is obtained by random initialization and model training optimization; L recon is the reconstruction loss, L equ is the constant variable loss, L sym is the symmetric loss, L EC is an equivariant constraint; λ1 and λ2 are the weight coefficients of equivariant loss and symmetric loss respectively; l represents the number of dimensions of each module; It means that the feature map of the i-th module has only the m-th dimension as 1 and the other dimensions as 0. It indicates that the feature map of the i-th module has only the n-th dimension as 1 and the other dimensions as 0; For arrive The optimal transformation parameters of The prediction matrix M is parameterized by δ (i) The value of the mth row and nth column of (δ); is the prediction matrix M with parameter δ′ (i) The value of the mth row and nth column of (δ′); T is a hyperparameter greater than 0; The transformation parameters are The prediction matrix when ; δ′ is the integral variable, which is used to distinguish the transformation parameter δ under different integral terms.
[0018] Preferably, the transformation comprises a translation transformation or a translation-rotation transformation.
[0019] The second aspect of the present invention provides a modular network autoencoder based on equivariant constraints, which is trained using the method described in the first aspect.
[0020] The third aspect of the present invention provides the application of the modular network autoencoder based on equivariant constraints as described in the second aspect in classification, detection, segmentation and associative generation.
[0021] A fourth aspect of the present invention provides a training device for a modular network autoencoder based on equivariant constraints, comprising:
[0022] A training data acquisition module, used to acquire training data, wherein the training data includes an image I and an image I′ obtained after the image I is transformed;
[0023] A training module is used to train the convolutional network autoencoder based on the training data using equivariant constraints to obtain a trained autoencoder; wherein the equivariant constraint L EC The calculation is performed using the following formula:
[0024]
[0025] L equ =∑ i ||f′ (i) -M (i) (δ)f (i) || 2 , i∈{0, 1, ..., k-1},
[0026]
[0027] L EC =L recon +λ1L equ +λ2L sym ,
[0028] f=I*W,
[0029] f′=I′*W,
[0030] Where I′=L δ (I), Lδ is a transformation with a transformation parameter δ; W is the convolution kernel of the convolutional network autoencoder, and W = [W (0) , W (1) , ..., W (k-1) ], that is, the convolution kernel is divided into k modules; f is the feature image obtained after the image I is convolved, and f = [f (0) , f (1) , ..., f (k-1) ]; f′ is the feature image obtained after convolution of image I′, f′=[f′ (0) , f′ (1) , ..., f′ (k-1) ]; * indicates convolution operation, represents the deconvolution operation; f (i) is the feature map obtained after image I is convolved by the i-th module of the convolution kernel; f′ (i) is the feature map obtained after the image I′ is convolved by the i-th module of the convolution kernel; M (i) (δ) is the prediction matrix of the i-th module with parameter δ, which is obtained by random initialization and model training optimization; L recon is the reconstruction loss, L equ is the constant variable loss, L sym is the symmetric loss, L EC is an equivariant constraint; λ1 and λ2 are the weight coefficients of equivariant loss and symmetric loss respectively; l represents the number of dimensions of each module; It means that the feature map of the i-th module has only the m-th dimension as 1 and the other dimensions as 0. It indicates that the feature map of the i-th module has only the n-th dimension as 1 and the other dimensions as 0; For arrive The optimal transformation parameters of The prediction matrix M is parameterized by δ (i) The value of the mth row and nth column of (δ); is the prediction matrix M with parameter δ′ (i) The value of the mth row and nth column of (δ′); T is a hyperparameter greater than 0; The transformation parameters are The prediction matrix when ; δ′ is the integral variable, which is used to distinguish the transformation parameter δ under different integral terms.
[0031] Preferably, the transformation comprises a translation transformation or a translation-rotation transformation.
[0032] A fifth aspect of the present invention provides a memory storing a plurality of instructions, wherein the instructions are used to implement the training method of the modular network autoencoder based on equivariant constraints as described in the first aspect.
[0033] The sixth aspect of the present invention provides an electronic device, comprising a processor and a memory connected to the processor, wherein the memory stores a plurality of instructions, which can be loaded and executed by the processor so that the processor can execute the training method of the modular network autoencoder based on equivariant constraints as described in the first aspect.
[0034] The beneficial effects of the present invention are as follows: the training method and device of the modular network autoencoder based on equivariant constraints provided by the present invention enhance the correlation within the autoencoder module and the independence between modules by constructing equivariant constraints and using equivariant constraints to train the modular network autoencoder on the training data of homologous image pairs, thereby promoting the specialization of network module functions. The functionally differentiated modular network is applied to downstream tasks such as classification, detection, segmentation and associative generation, which can provide a better representation space for downstream tasks and is conducive to improving the robustness and generalization of downstream tasks. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 The following is a flow chart of the training method of the modular network autoencoder based on equivariant constraints according to the present invention:
[0036] Figure 2 Schematic diagram of the training process of the modular network autoencoder;
[0037] Figure 3 It is a schematic diagram of the functional differentiation effect under modular equivariant constraints;
[0038] Figure 4 This is a functional structural diagram of a training device for a modular network autoencoder based on equivariant constraints according to the present invention. DETAILED DESCRIPTION
[0039] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0040] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a memory, and a display screen. The memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.
[0041] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory.
[0042] The memory may include random access memory (RAM) or read-only memory (ROM). The memory may be used to store instructions, programs, codes, code sets or instructions.
[0043] The display screen is used to display the user interface of each application.
[0044] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.
[0045] Embodiment 1
[0046] like Figure 1 As shown, an embodiment of the present invention provides a training method for a modular network autoencoder based on equivariant constraints, comprising:
[0047] S101, obtaining training data, wherein the training data includes an image I and an image I′ obtained after the image I is transformed;
[0048] S102, based on the training data, train the convolutional network autoencoder using equivariant constraints to obtain a trained autoencoder; wherein the equivariant constraint L EC The calculation is performed using the following formula:
[0049]
[0050] L equ =∑ i ||f′ (i) -M(i) (δ)f (i) || 2 , i∈{0, 1, ..., k-1},
[0051]
[0052] L EC =L recon +λ1L equ +λ2L sym ,
[0053] f=I*W,
[0054] f′=I′*W,
[0055] Where I′=L δ (I), Lδ is a transformation with a transformation parameter δ; W is the convolution kernel of the convolutional network autoencoder, and W = [W (0) , W (1) , ..., W (k-1) ], that is, the convolution kernel is divided into k modules; f is the feature image obtained after the image I is convolved, and f = [f (0) , f (1) , ..., f (k-1) ]; f′ is the feature image obtained after convolution of image I′, f′=[f′ (0) , f′ (1) , ..., f, (k-1) ]; * indicates convolution operation, represents the deconvolution operation; f (i) is the feature map obtained after image I is convolved by the i-th module of the convolution kernel; f′ (i) is the feature map obtained after the image I′ is convolved by the i-th module of the convolution kernel; M (i) (δ) is the prediction matrix of the i-th module with parameter δ, which is obtained by random initialization and model training optimization; L recon is the reconstruction loss, L equ is the constant variable loss, L sym is the symmetric loss, L EC is an equivariant constraint; λ1 and λ2 are the weight coefficients of equivariant loss and symmetric loss respectively; l represents the number of dimensions of each module; It means that the feature map of the i-th module has only the m-th dimension as 1 and the other dimensions as 0. It indicates that the feature map of the i-th module has only the n-th dimension as 1 and the other dimensions as 0; For arrive The optimal transformation parameters of The prediction matrix M is parameterized by δ (i)The value of the mth row and nth column of (δ); is the prediction matrix M with parameter δ′ (i) The value of the mth row and nth column of (δ′); T is a hyperparameter greater than 0; The transformation parameters are The prediction matrix when ; δ′ is the integral variable, which is used to distinguish the transformation parameter δ under different integral terms. It should be noted that The subscript "m" in The subscript "m" in , although it represents the number of dimensions in the feature map and the number of rows in the matrix, but since the feature map and the matrix are corresponding, The subscript "m" in The subscript "m" in is corresponding. In order to illustrate the corresponding relationship between the two, the same letter "m" is used to represent them in the present invention. Similarly, The subscript "n" in The subscript "n" in , although it represents the number of dimensions in the feature map and the number of columns in the matrix, but since the feature map and the matrix are corresponding, The subscript "n" in The subscript "n" in is corresponding. In order to illustrate the corresponding relationship between the two, the same letter "n" is used to represent them in the present invention.
[0056] In the above training method, the training process of the modular network autoencoder can be as follows Figure 2 shown.
[0057] In a preferred embodiment of the present invention, in step S101, the transformation may include a translation transformation or a translation rotation transformation, so as to obtain a homologous data pair image I and image I'.
[0058] In the actual application process of the present invention, an autoencoder composed of convolution can be constructed first. And the convolution kernel is divided into k modules. When training the autoencoder, the untransformed image I and the transformed image I′ can be simultaneously input into the constructed network, and the k modules are used to perform convolution operations respectively to obtain the feature image corresponding to each module. In addition, an equivariant constraint condition can be constructed, and the equivariant constraint condition can be used to train and optimize the convolution kernel W and the prediction matrix M of the constructed convolution network autoencoder. After the training is completed, all parameters are solidified to obtain a modular network with differentiated functions, which can be used for downstream tasks such as recognition and associative generation.
[0059] By constructing modular equivariant constraints, the present invention proposes a training method for modular autoencoders. The equivariant constraints are used to enhance the correlation within the modules and the independence between the modules, so that the modules can form spontaneous functional differentiation. Figure 3The effect of functional differentiation under modular equivariant constraints is demonstrated. Figure 3 In the figure, each row in Figure a, Figure b and Figure c is a module, and each block is a visualization of a learned convolution kernel. It can be seen from the figure that there is an obvious functional differentiation between the modules, such as differentiation of different spatial frequencies, brightness and color. Figure d shows the visualization of the reconstruction of the features extracted by each module in Figure c, and the differentiation of the functions between the modules in terms of color, brightness and edges of different spatial frequencies can be more intuitively seen. This differentiated modular feature is conducive to providing a better representation space for downstream tasks, such as classification tasks, detection tasks and associative generation tasks. At the same time, it is conducive to improving the robustness and generalization of downstream tasks. For example, when the input image style of the recognition task changes greatly, the conventional classification network may be unable to recognize the phenomenon, that is, the generalization ability is poor. However, for the modular representation in the present invention, even if the image style changes, it will only affect the representation of a few modules, and will not have much impact on other modules. Therefore, it can still be better recognized through integrated learning, thereby improving the generalization ability of the network.
[0060] Embodiment 2
[0061] like Figure 4 As shown, another aspect of the present invention also includes a functional module architecture that is completely consistent with the aforementioned method flow, that is, an embodiment of the present invention also provides a training device for a modular network autoencoder based on equivariant constraints, including:
[0062] A training data acquisition module 301 is used to acquire training data, wherein the training data includes an image I and an image I′ obtained after the image I is transformed;
[0063] The training module 302 is used to train the convolutional network autoencoder based on the training data using the equivariant constraint to obtain a trained autoencoder; wherein the equivariant constraint L EC The calculation is performed using the following formula:
[0064]
[0065] L equ =∑ i ||f′ (i) -M (i) (δ)f (i) || 2 , i∈{0, 1, ..., k-1},
[0066]
[0067] L EC =L recon +λ1L equ +λ2Lsym ,
[0068] f=I*W,
[0069] f′=I′*W,
[0070] Where I′=L δ (I), Lδ is a transformation with a transformation parameter δ; W is the convolution kernel of the convolutional network autoencoder, and W = [W (0) , W (1) , ..., W (k-1) ], that is, the convolution kernel is divided into k modules; f is the feature image obtained after the image I is convolved, and f = [f (0) , f (1) , ..., f (k-1) ]; f′ is the feature image obtained after convolution of image I′, f′=[f′ (0) , f′ (1) , ..., f′ (k-1) ]; * indicates convolution operation, represents the deconvolution operation; f (i) is the feature map obtained after image I is convolved by the i-th module of the convolution kernel; f′ (i) is the feature map obtained after the image I′ is convolved by the i-th module of the convolution kernel; M (i) (δ) is the prediction matrix of the i-th module with parameter δ, which is obtained by random initialization and model training optimization; L recon is the reconstruction loss, L equ is the constant variable loss, L sym is the symmetric loss, L EC is an equivariant constraint; λ1 and λ2 are the weight coefficients of equivariant loss and symmetric loss respectively; l represents the number of dimensions of each module; It means that the feature map of the i-th module has only the m-th dimension as 1 and the other dimensions as 0. It indicates that the feature map of the i-th module has only the n-th dimension as 1 and the other dimensions as 0; For arrive The optimal transformation parameters of The prediction matrix M is parameterized by δ (i) The value of the mth row and nth column of (δ); is the prediction matrix M with parameter δ′ (i) The value of the mth row and nth column of (δ′); T is a hyperparameter greater than 0; The transformation parameters are The prediction matrix when ; δ′ is the integral variable, which is used to distinguish the transformation parameter δ under different integral terms.
[0071] The device can be implemented through the training method of the modular network autoencoder based on equivariant constraints provided in the above-mentioned embodiment 1. The specific implementation method can be found in the description of embodiment 1 and will not be repeated here.
[0072] The present invention also provides a memory storing a plurality of instructions, wherein the instructions are used to implement the training method of the modular network autoencoder based on equivariant constraints as described in Example 1.
[0073] The present invention also provides an electronic device, comprising a processor and a memory connected to the processor, wherein the memory stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the training method of the modular network autoencoder based on equivariant constraints as described in Example 1.
[0074] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A training method for a modular network autoencoder based on equivariant constraints, characterized in that: include: Acquire training data, wherein the training data includes an image I and an image I′ obtained after the image I is transformed; Based on the training data, the convolutional network autoencoder is trained using equivariant constraints to obtain a trained autoencoder; wherein the equivariant constraint L EC The calculation is performed using the following formula: L equ =∑ i ||f′ (i) -M (i) (δ)f (i) || 2 ,i∈{0,1,...,k-1}, L EC =L recon +λ1L equ +λ2L sym , f=I*W, f′=I′*W, Where I′=L δ (I), Lδ is a transformation with a transformation parameter δ; W is the convolution kernel of the convolutional network autoencoder, and W = [W (0) , W (1) , ..., W (k-1) ], that is, the convolution kernel is divided into k modules; f is the feature image obtained after the image I is convolved, and f = [f (0) , f (1) , ..., f (k-1) ]; f′ is the feature image obtained after convolution of image I′, f′=[f′ (0) , f′ (1) , ..., f′ (k-1) ]; * indicates convolution operation, represents the deconvolution operation; f (i) is the feature map obtained after image I is convolved by the i-th module of the convolution kernel; f′ (i) is the feature map obtained after the image I′ is convolved by the i-th module of the convolution kernel; M (i) (δ) is the prediction matrix of the i-th module with parameter δ, which is obtained by random initialization and model training optimization; L recon is the reconstruction loss, L equ is the constant variable loss, L sym is the symmetric loss, L EC is an equivariant constraint; λ1 and λ2 are the weight coefficients of equivariant loss and symmetric loss respectively; l represents the number of dimensions of each module; It means that the feature map of the i-th module has only the m-th dimension as 1 and the other dimensions as 0. It indicates that the feature map of the i-th module has only the n-th dimension as 1 and the other dimensions as 0; For arrive The optimal transformation parameters of The prediction matrix M is parameterized by δ (i) The value of the mth row and nth column of (δ); is the prediction matrix M with parameter δ′ (i) The value of the mth row and nth column of (δ′); T is a hyperparameter greater than 0; The transformation parameters are The prediction matrix when ; δ′ is the integral variable, which is used to distinguish the transformation parameter δ under different integral terms.
2. The training method of modular network autoencoder based on equivariant constraints according to claim 1, characterized in that: The transformation includes a translation transformation or a translation-rotation transformation.
3. A modular network autoencoder based on equivariant constraints, characterized in that: The method is trained by any one of claims 1 to 2.
4. Application of the modular network autoencoder based on equivariant constraints as described in claim 3 in classification, detection, segmentation and associative generation.
5. A training device for a modular network autoencoder based on equivariant constraints, characterized in that: include: A training data acquisition module, used to acquire training data, wherein the training data includes an image I and an image I′ obtained after the image I is transformed; A training module is used to train the convolutional network autoencoder based on the training data using equivariant constraints to obtain a trained autoencoder; wherein the equivariant constraint L EC The calculation is performed using the following formula: L equ =∑ i ||f′ (i) -M (i) (δ)f (i) || 2 ,i∈{0,1,...,k-1}, L EC =L recon +λ1L equ +λ2L sym , f=I*W, f′=I′*W, Where I′=L δ (I), Lδ is a transformation with a transformation parameter δ; W is the convolution kernel of the convolutional network autoencoder, and W = [W (0) , W (1) , ..., W (k-1) ], that is, the convolution kernel is divided into k modules; f is the feature image obtained after the image I is convolved, and f = [f (0) , f (1) , ..., f (k-1) ]; f′ is the feature image obtained after convolution of image I′, f′=[f′ (0) , f′ (1) , ..., f′ (k-1) ]; * indicates convolution operation, represents the deconvolution operation; f (i) is the feature map obtained after image I is convolved by the i-th module of the convolution kernel; f' (i) is the feature map obtained after the image I′ is convolved by the i-th module of the convolution kernel; M (i) (δ) is the prediction matrix of the i-th module with parameter δ, which is obtained by random initialization and model training optimization; L recon is the reconstruction loss, L equ is the constant variable loss, L sym is the symmetric loss, L EC is an equivariant constraint; λ1 and λ2 are the weight coefficients of equivariant loss and symmetric loss respectively; l represents the number of dimensions of each module: It means that the feature map of the i-th module has only the m-th dimension as 1 and the other dimensions as 0. It indicates that the feature map of the i-th module has only the n-th dimension as 1 and the other dimensions as 0; For arrive The optimal transformation parameters of The prediction matrix M is parameterized by δ (i) The value of the mth row and nth column of (δ); is the prediction matrix M with parameter δ′ (i) The value of the mth row and nth column of (δ′); T is a hyperparameter greater than 0; The transformation parameters are The prediction matrix when ; δ′ is the integral variable, which is used to distinguish the transformation parameter δ under different integral terms.
6. The training device of modular network autoencoder based on equivariant constraints according to claim 5, characterized in that: The transformation includes a translation transformation or a translation-rotation transformation.
7. A memory, characterized in that: Multiple instructions are stored, and the instructions are used to implement the training method of the modular network autoencoder based on equivariant constraints as described in any one of claims 1-2.
8. An electronic device, characterized in that: It includes a processor and a memory connected to the processor, the memory stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the training method of the modular network autoencoder based on equivariant constraints as described in any one of claims 1-2.
Citation Information
Patent Citations
Gating feature attention equivariant segmentation method based on weak supervised learning
CN114373094A
OCT (Optical Coherence Tomography) super-resolution reconstruction method and device based on isovariant learning and prior guidance
CN116128728A
