A contribution evaluation method and system based on value gain in federated learning

Abstract

The present invention provides a method for value-gain-based contribution assessment in federated learning, comprising: step S1. calculating the correlation between each client's local model and the global model in round t based on a contribution assessment metric for correlation between multidimensional computing models; step S2. calculating the parameter distance between the global model and the preheating model in round t based on a dynamic adjustment mechanism and performing a normalization operation to obtain the contribution weight of each participating client in each communication round; and step S3. calculating the client's true contribution value based on the correlation between each client's local model and the global model in round t in step 1 and the contribution weight of each participating client in each communication round in step 2. In the field of client contribution assessment, the present method effectively reduces computational complexity and is independent of a test dataset.

Description

Technical Field

[0001] The present invention relates to the field of federated learning technology, and in particular to a method and system for evaluating contribution based on value gain in federated learning. Background Art

[0002] Existing methods for assessing customer contributions primarily include schemes based on heuristic Shapley values ​​and parameter correlation. Heuristic Shapley value-based methods assess contribution by calculating each participant's marginal contribution to the collective value, taking into account factors such as data volume, test accuracy, and data diversity. While theoretically attractive, their computational complexity makes them unsuitable for large-scale customer participation. Another approach is based on model parameter correlation, which calculates the distance between local and global gradients; the closer the distance, the greater the customer's contribution. These methods significantly reduce computational complexity and eliminate the reliance on test datasets. However, compared to centralized training, current parameter correlation-based methods ignore the key characteristics of federated learning. Specifically, the global model gradually converges through iterative communication rounds, each of which requires the participation of a large number of customers. Existing methods only consider the positive impact of participants on the collective, while ignoring the performance gains of the global model compared to previous rounds. This oversight results in a situation where participants appear to have made significant contributions despite negligible collective gains, hindering the accurate assessment of each customer's true contribution. For example, even if the performance gain of the global model is almost zero, the correlation between the local and global model parameters still exists, which can still be regarded as a contribution and may even appear stronger than the previous communication round.

[0003] In order to maintain the fairness of the algorithm while reducing communication costs, the present invention fully studies the performance improvement of the global model in each communication round while controlling the computational complexity. The client's contribution is finely adjusted according to the performance gain of the global model in each communication round, thereby improving the fairness of the evaluation method. However, there is a difficulty in implementing this method: how to quantify the performance gain of each round of communication of the global model while managing resource overhead. This function determines whether the evaluation algorithm can finely evaluate the contribution of each client in each round of communication. Currently, there is a lack of a unified quantitative benchmark for evaluating model performance, and most methods rely on test datasets and a large amount of computing resources. Therefore, it is necessary to design a method to quantify the performance gain of the global model while minimizing resource overhead. Summary of the Invention

[0004] This invention provides a method and system for evaluating contribution based on value gain in federated learning. Without introducing additional computing resources or test datasets, the change in weight parameters is used as an indicator to measure model performance gain. When evaluating a client's contribution, the contribution index for the current round of communication is first calculated based on the correlation between the local model and the global model parameters. The distance between the global model and the pre-aggregated model after each round is then calculated as the performance gain of the global model. The contribution of each client in each communication round is dynamically adjusted based on the global model performance gain.

[0005] In order to solve the above problems, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a contribution evaluation method based on value gain in federated learning, comprising:

[0007] Step S1. Calculate the correlation between each client's local model and the global model in round t based on the contribution evaluation index of the correlation between the multi-dimensional computing models;

[0008] Step S2. Calculate the parameter distance between the global model and the preheating model in round t based on the dynamic adjustment mechanism and perform normalization to obtain the contribution weight of each communication round of the participating clients;

[0009] Step S3. Calculate the client's true contribution value based on the correlation between each client's local model and the global model in the tth round in step 1 and the contribution weight of each client participating in each communication round in step 2.

[0010] Furthermore, the correlation between each client's local model and the global model in the tth round in step 1 is calculated as follows:

[0011]

[0012] Among them, sim(i) t represents the correlation between the local model of client i and the global model in round t, represents the parameter matrix of the local model of client i in the tth round, W t+1 represents the parameter matrix of the global model at the t+1th round, num is the scale of the model parameters; |||| represents the second norm; ∑ represents the sum.

[0013] Furthermore, step 2 includes the following sub-steps:

[0014] S21. Pre-training: Pre-train the model based on the initialization model provided by the central server when starting federated training to obtain a warm-up model;

[0015] S22. Model Training: The server sends the initialized global model, also known as the warm-up model. Each participant receives the global model and trains on it using their local dataset, obtaining updated local models and uploading them to the central server. The central server then aggregates the updated local models from each participating client into a new global model. This process is repeated until the global model converges or a predefined number of training rounds is reached.

[0016] S23. Contribution weight calculation: Quantify the changes in the global model using the preheated model as a benchmark. Calculate the parameter distance between the global model and the preheated model in round t and perform normalization operations to obtain the contribution weight of each participating client in each communication round.

[0017] Furthermore, the preheating model in step S21 is:

[0018]

[0019] Among them, W pre is the warm-up model parameter matrix, W init is the initialization model parameter matrix given by the central server when starting federated training, η is the learning rate, N is the total number of clients, and n is the number of warm-up rounds. is the warm-up training loss function, where is the parameter matrix of the local model of client i in round t, ∑ represents the sum, D i is the number of local data samples held by client i.

[0020] Furthermore, the step S22 includes:

[0021] (1) Initialization of global model and local model

[0022] At the beginning of the tth round of training in federated learning, the central server maintains a global model. The initial value of the local model of each client i in each round is the parameter value of the global model in the previous round:

[0023] W pre =W 0

[0024]

[0025] Among them, W pre is the warm-up model parameter matrix, W 0 is the initial value of the global model; W t-1 is the global model parameter matrix for round t-1, is the parameter matrix of the local model of client i in round t;

[0026] (2) Local training process: gradient descent update

[0027] After receiving the global model parameters, each client uses its own local data to train the model. Each client updates its model parameters through gradient descent:

[0028]

[0029] in, is the parameter matrix of the local model of client i in the t+1th round, η is the learning rate, represents the loss function of client i in round t, is the loss function with respect to the parameter gradient;

[0030] By continuously performing gradient updates, client i improves the local model parameters based on its local data;

[0031] (3) Local model aggregation:

[0032] After local training is completed, each client sends its updated model parameters back to the central server. The central server aggregates the updates from multiple participating clients into a new global model through the federated averaging method. The update formula of the aggregated global model is:

[0033]

[0034] Where m is the number of clients participating in the training in the current round, N is the total number of clients, and W t+1 is the global model parameter matrix for round t+1.

[0035] Furthermore, the loss function of client i in round t is:

[0036]

[0037] in, The loss function calculates the loss between the sample prediction value and the true value during the local client training process. W represents the global model received by the client. is the predicted value of the parameter matrix of the local model of client i in the tth round when the input is x, and y is the true value of the parameter matrix of the local model of client i in the tth round when the input is x.

[0038] Furthermore, the contribution weight φ(t) of each communication round of the participating clients in step S23 is:

[0039]

[0040] Among them, cor(W t ,W pre ) is the global model W of the tth roundt With preheating model W pre The distance between parameters;

[0041]

[0042] Among them, k is the subscript of the model parameter matrix, num is the model parameter scale, is the kth element of the global model parameter matrix of the tth round, is the kth element of the warmup model parameter matrix.

[0043] Furthermore, the actual contribution value con(i) of the client in step S3 is:

[0044]

[0045] Among them, sim(i) t represents the correlation between the local model of client i and the global model in the tth round, φ(t) is the contribution weight of each participating client in each communication round, m is the total number of participating clients in each round, T is the total number of iteration rounds, and ∑ represents the sum.

[0046] In another aspect, the present invention provides a contribution evaluation system based on value gain in federated learning, comprising:

[0047] The first module is used to calculate the correlation between each client's local model and the global model in the tth round based on the contribution evaluation index of the correlation between the multi-dimensional computing models;

[0048] The second module is used to calculate the parameter distance between the global model and the preheating model in the tth round based on the dynamic adjustment mechanism and perform normalization operation to obtain the contribution weight of each communication round of the participating clients;

[0049] The third module is used to calculate the true contribution value of the client based on the correlation between the local model of each client and the global model in the tth round and the contribution weight of the participating clients in each communication round.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. In the field of client contribution evaluation, this method effectively reduces the computational complexity and does not depend on the test data set.

[0052] 2. When a small number of clients participate, the comparative benchmark experiment proves that the present invention still has a fairer evaluation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 A flow chart of a contribution evaluation method based on value gain in federated learning provided by an embodiment of the present invention;

[0055] Figure 2 Schematic diagram showing the comparison of contributions calculated by FedCA using MNIST in different scenarios in Example 1 of the present invention;

[0056] Figure 3 Schematic diagram comparing contributions calculated by FedCA using Fashion-MNIST in different scenarios in Example 1 of the present invention;

[0057] Figure 4 Schematic diagram comparing the contribution calculated by FedCA using CIFAR-10 in different scenarios in Example 1 of the present invention. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0059] It should be noted that, in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include a ..." do not exclude the presence of other identical elements in the process, method, article or device comprising the elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0060] The terms "first," "second," and the like in this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that such terms are interchangeable where appropriate, so that embodiments of this application can be implemented in an order other than that illustrated or described herein. Furthermore, the terms "first," "second," and the like generally distinguish objects of a class and do not limit the number of objects; for example, the first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the connected objects.

[0061] Example 1

[0062] The following combination Figure 1 The contribution evaluation method based on value gain in federated learning provided by an embodiment of the present invention is described, including:

[0063] Step S1. Calculate the correlation between each client's local model and the global model in round t based on the contribution evaluation index of the correlation between the multi-dimensional computing models;

[0064] Single-dimensional measurement methods (such as cos, -norm, etc.) to calculate the correlation between local and global model parameters. The loss of correlation information between model parameters is too serious. The cosine similarity between weights is difficult to represent the correlation between models, which leads to other problems such as inaccurate evaluation.

[0065] In this implementation step S1, a contribution evaluation index sim(i) for the correlation between multi-dimensional computing models is proposed, which strives to accurately calculate the correlation between models and improve the accuracy of the evaluation. The current federated learning training mechanism is manifested in that multiple clients use the local data they hold to train local models. The parameter matrices of the local and global models continuously change with model training. From the process point of view, the parameter matrix is ​​a high-dimensional vector. It can be abstracted that each client participating in the aggregation contributes a thrust to the global model in the high-dimensional solution space. This thrust is not only important in terms of the pushing distance, but also in terms of the pushing direction. Therefore, the calculation method of the correlation between each client's local model and the global model in the tth round in step 1 is;

[0066]

[0067] Among them, sim(i) t represents the correlation between the local model of client i and the global model in round t, represents the parameter matrix of the local model of client i in the tth round, W t+1represents the parameter matrix of the global model at round t+1, num is the scale of the model parameters, |||| represents the binorm, and ∑ represents the sum. This factor comprehensively considers the similarity between the local and global optimization directions and the distance between the optimal solutions. A greater similarity indicates a closer fit between the optimization directions of the local and global models.

[0068] Step S2. Calculate the parameter distance between the global model and the preheating model in round t based on the dynamic adjustment mechanism and perform normalization to obtain the contribution weight of each communication round of the participating clients;

[0069] The core of the contribution evaluation algorithm in this embodiment is to dynamically increase each client's contribution based on the performance gain of the global model as communication rounds iterate. Traditional model performance evaluation schemes rely on metrics such as accuracy, recall, F1 score, and AUC-ROC on a validation set, which require additional datasets and computing resources. To reduce this overhead, we can evaluate performance gains by observing the magnitude of changes in model weight parameters. This approach effectively reflects the optimization effect of the model without introducing additional datasets and reduces computational costs.

[0070] Specifically include:

[0071] S21. Pre-training: Pre-train the model based on the initialization model provided by the central server when starting federated training to obtain a warm-up model;

[0072] The contributions of participants in the current communication round are not only perceived by the aggregated global model, but also by the performance improvement of the aggregated global model compared to the global model in the previous round. First, we perform specific pre-training to perceive all datasets. The goal is to roughly determine the global optimization direction and the goal of the global optimal solution in the solution space.

[0073] The preheating model in step S21 is:

[0074]

[0075] Among them, W pre is the warm-up model parameter matrix, W init is the initialization model parameter matrix given by the central server when starting federated training, η is the learning rate, N is the total number of clients, and n is the number of warm-up rounds. is the warm-up training loss function, where is the parameter matrix of the local model of client i in round t, ∑ represents the sum, D i is the number of local data samples held by client i.

[0076] S22. Model Training: The server sends the initialized global model, also known as the warm-up model. Each participant receives the global model and trains on it using their local dataset, obtaining updated local models and uploading them to the central server. The central server then aggregates the updated local models from each participating client into a new global model. This process is repeated until the global model converges or a predefined number of training rounds is reached.

[0077] Step S22 includes:

[0078] (1) Initialization of global model and local model

[0079] At the beginning of the tth round of training in federated learning, the central server maintains a global model. The initial value of the local model of each client i in each round is the parameter value of the global model in the previous round:

[0080] W pre =W 0

[0081]

[0082] Among them, W pre is the warm-up model parameter matrix, W 0 is the initial value of the global model; W t-1 is the global model parameter matrix for round t-1, is the parameter matrix of the local model of client i in round t;

[0083] (2) Local training process: gradient descent update

[0084] After receiving the global model parameters, each client uses its own local data to train the model. Each client updates its model parameters through gradient descent. The goal is to train the model by minimizing the loss function:

[0085]

[0086] in, is the parameter matrix of the local model of client i in the t+1th round, η is the learning rate, represents the loss function of client i in round t, is the loss function with respect to the parameter gradient;

[0087] The loss function of client i in round t is:

[0088]

[0089] in, The loss function calculates the loss between the sample prediction value and the true value during the local client training process. W represents the global model received by the client. is the predicted value of the parameter matrix of the local model of client i in the tth round when the input is x, and y is the true value of the parameter matrix of the local model of client i in the tth round when the input is x.

[0090] By continuously performing gradient updates, client i improves the local model parameters based on its local data;

[0091] (3) Local model aggregation:

[0092] After local training is completed, each client sends its updated model parameters back to the central server. The central server aggregates the updates from multiple participating clients into a new global model through the federated averaging method. The update formula of the aggregated global model is:

[0093]

[0094] Where m is the number of clients participating in the training in the current round, N is the total number of clients, and W t+1 is the global model parameter matrix for round t+1.

[0095] S23. Contribution weight calculation: Quantify the changes in the global model using the preheated model as a benchmark. Calculate the parameter distance between the global model and the preheated model in round t and perform normalization operations to obtain the contribution weight of each participating client in each communication round.

[0096] The contribution weight φ(t) of each participating client in each communication round in step S23 is:

[0097]

[0098] Among them, cor(W t ,W pre ) is the global model W of the tth round t With preheating model W prd The distance between parameters;

[0099]

[0100] Among them, k is the subscript of the model parameter matrix, num is the model parameter scale, is the kth element of the global model parameter matrix of the tth round, is the kth element of the warmup model parameter matrix.

[0101] This weighted average algorithm ensures that clients with more data contribute more to the global model update, thereby effectively reflecting the differences in data distribution and ensuring that the global model can better adapt to the data of all clients.

[0102] Step S3. Calculate the client's true contribution value based on the correlation between each client's local model and the global model in the tth round in step 1 and the contribution weight of each client participating in each communication round in step 2.

[0103] Based on the above theoretical basis, we observe that the performance of the global model is almost monotonically increasing. The parameter matrix of the global model represents the optimal solution and optimization direction of the training process. Therefore, the greater the distance between the weight parameters of the global models, the greater the change in the global model. The improvement in global model performance is almost monotonically increasing, which leads to the deduction that the global model has greater progress in the performance curve.

[0104] The actual contribution value con(i) of the client in step S3 is:

[0105]

[0106] Among them, sim(i) t represents the correlation between the local model of client i and the global model in the tth round, φ(t) is the contribution weight of each participating client in each communication round, m is the total number of participating clients in each round, T is the total number of iteration rounds, and ∑ represents the sum.

[0107] WPre is a warm-up model, which aims to roughly determine the optimal solution in the parameter space under all data and use W pre quantify the changes of the global model to the benchmark, that is, the learning of the global model on each local data feature. cor(W t , W pre ) represents the parameter distance between the global model of round t and the preheated model. φ(t) represents the contribution weight assigned to the training of round t after normalization, indicating how much the current round of training helps improve the performance of the global model. We quantified the performance gain of the global model in each round, and based on this, proposed a dynamic adjustment mechanism φ(t) to fine-grainedly evaluate the contribution of participants in each communication round. In summary, the FedCA algorithm uses the parameter correlation between multi-dimensional computing models as the basic indicator to measure contribution (the degree of closeness between the local optimum of the current round and the global optimum of the current round). Through the dynamic adjustment mechanism (the distance between the global optimum of the current round and the global optimal solution), it calculates the true driving force of the local contribution parameters on the global model approaching the global optimal solution, accurately evaluates the contribution of the client, and can adapt to the complex training environment of FL.

[0108] The performance of the FedCA algorithm was validated using various neural network architectures and datasets. A convolutional neural network (CNN) was used for the MNIST and Dashonin-MNIST datasets, and a residual network (ResNet) was used for the CIFAR-10 dataset. The detailed experimental setup parameters are listed in Table 1 below.

[0109] TABLE 1.DETAILS OF PARAMETERS.

[0110]

[0111] In the image classification task, to comprehensively evaluate the FedCA algorithm's sensitivity to abnormalities in user data, we carefully designed the following four experimental scenarios: 1) normal scenario, 2) impulse noise scenario, 3) Gaussian noise scenario, and 4) label confusion scenario. By running the FedCA algorithm in these scenarios, we calculated the contribution of each client participating in FL on three classic image classification datasets: MNIST, Fashion-MNIST, and CIFAR-10. Figure 2-Figure 4 Figure 2 shows a comparison of the contributions calculated by FedCA under different scenarios, visualizing the algorithm's sensitivity to various data anomalies. In the FL experiments, we first randomly and equally distribute the dataset to 10 local clients without any specific treatment, ensuring that all participate in each training round with a fixed participation ratio of 1.0. Using the FedCA algorithm, we find uniform client contributions, with each client contributing approximately 10%, in standard training rounds on the MNIST, FastMNIST, and CIFAR-10 datasets. To further explore the effectiveness of the algorithm, we respond to data quality variations by introducing impulse noise (random black and white dots) and Gaussian noise (random brightness variations) to the datasets of the last two clients. In the experiments, a small amount of impulse noise was added to the datasets of the last two clients to simulate image quality degradation caused by sensor failure or signal transmission errors in real applications. The results show that these clients receive lower contribution estimates, demonstrating FedCA's sensitivity to data quality and its ability to distinguish between minor and more complex noise effects. Finally, in an experiment simulating label confusion caused by mislabeling of the last two customers’ data, normal data produced a higher contribution rate, while interrupted data significantly reduced customer contribution, confirming the effectiveness of FedCA in identifying and assessing the quality and accuracy of data provided by each customer.

[0112] In a comparative test of contribution evaluation on the federated learning image classification task, three different experimental scenarios were designed. These scenarios include 1) a normal scenario, i.e., a standard FL setting; 2) a random noise scenario, in which random noise is introduced to simulate data contamination; and 3) a mislabeled scenario, used to test the robustness of the algorithm in the presence of mislabeled data. To verify the accuracy of the FedCA algorithm in federated contribution evaluation, its performance was compared with two existing methods: the game-theoretic Shapley value method and the FedCM algorithm. While theoretically rigorous, the Shapley value method requires significant computational resources and offline calculations. Comparability of the experimental data was ensured by maintaining consistent conditions across all methods. Detailed results are shown in the table below, highlighting the performance of the FedCA algorithm in the three different scenarios.

[0113]

[0114] Example 2

[0115] This embodiment provides a contribution evaluation system based on value gain in federated learning, including:

[0116] The first module is used to calculate the correlation between each client's local model and the global model in the tth round based on the contribution evaluation index of the correlation between the multi-dimensional computing models;

[0117] The second module is used to calculate the parameter distance between the global model and the preheating model in the tth round based on the dynamic adjustment mechanism and perform normalization operation to obtain the contribution weight of each communication round of the participating clients;

[0118] The third module is used to calculate the true contribution value of the client based on the correlation between the local model of each client and the global model in the tth round and the contribution weight of the participating clients in each communication round.

[0119] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A contribution evaluation method based on value gain in federated learning, characterized by: include: Step S1. Calculate the correlation between each client's local model and the global model in round t based on the contribution evaluation index of the correlation between the multi-dimensional computing models; The correlation between each client's local model and the global model in round t in step 1 is calculated as follows; in, Represents the client i The correlation between the local model and the global model at round t, Represents the client i The parameter matrix of the local model at the tth round, represents the parameter matrix of the global model at the t+1th round, is the scale of the model parameters; represents the two-norm; ∑ represents the sum; Step S2. Calculate the parameter distance between the global model and the preheating model in round t based on the dynamic adjustment mechanism and perform normalization to obtain the contribution weight of each communication round of the participating clients; the preheating model is: in, is the warm-up model parameter matrix, is the initialization model parameter matrix given by the central server when starting federated training, η is the learning rate, N is the total number of clients, and n is the number of warm-up rounds. is the warm-up training loss function, where For the client i The parameter matrix of the local model in the tth round, ∑ represents the sum, is the number of local data samples held by client i; Step S3. Calculate the client's true contribution value based on the correlation between each client's local model and the global model in the tth round in step 1 and the contribution weight of each client participating in each communication round in step 2.

2. The method for evaluating contribution based on value gain in federated learning according to claim 1, characterized in that: The step 2 includes the following sub-steps: S21. Pre-training: Pre-train the model based on the initialization model provided by the central server when starting federated training to obtain a warm-up model; S22. Model Training: The server sends the initialized global model, also known as the warm-up model. Each participant receives the global model and trains on it using their local dataset, obtaining updated local models and uploading them to the central server. The central server then aggregates the updated local models from each participating client into a new global model. This process is repeated until the global model converges or a predefined number of training rounds is reached. S23. Contribution weight calculation: Quantify the changes in the global model using the preheated model as a benchmark. Calculate the parameter distance between the global model in round t and the preheated model and perform a normalization operation to obtain the contribution weight of each participating client in each communication round.

3. The value-gain-based contribution evaluation method in federated learning according to claim 2, characterized in that: The step S22 includes: (1) Initialization of global model and local model At the beginning of the tth round of training in federated learning, the central server maintains a global model. The initial value of the local model of each client i in each round is the parameter value of the global model in the previous round: in, is the warm-up model parameter matrix, is the initial value of the global model; is the global model parameter matrix for round t-1, is the parameter matrix of the local model of client i in round t; (2) Local training process: gradient descent update After receiving the global model parameters, each client uses its own local data to train the model. Each client updates its model parameters through gradient descent: in, For the client i The parameter matrix of the local model in the t+1th round, η is the learning rate, represents the loss function of client i in round t, is the loss function with respect to the parameter gradient; By continuously performing gradient updates, client i improves the local model parameters based on its local data; (3) Local model aggregation: After local training is completed, each client sends its updated model parameters back to the central server. The central server aggregates the updates from multiple participating clients into a new global model through the federated averaging method. The update formula of the aggregated global model is: Where m is the number of clients participating in the training in the current round, N is the total number of clients, is the global model parameter matrix for round t+1.

4. The method for evaluating contribution based on value gain in federated learning according to claim 3, wherein: The loss function of client i in round t is: in, The loss function calculates the loss between the sample prediction value and the true value during the local client training process. W represents the global model received by the client. is the predicted value of the parameter matrix of the local model of client i in the tth round when the input is x, and y is the true value of the parameter matrix of the local model of client i in the tth round when the input is x.

5. The method for evaluating contribution based on value gain in federated learning according to claim 4, wherein: The contribution weight of each communication round of the participating clients in step S23 for: in, is the global model of round t With preheating model The distance between parameters; in, is the model parameter matrix subscript, is the model parameter scale, is the kth element of the global model parameter matrix of the tth round, is the kth element of the warmup model parameter matrix.

6. The method for evaluating contribution based on value gain in federated learning according to claim 1, wherein: The actual contribution value of the client in step S3 for: in, represents the correlation between the local model of client i and the global model in round t, is the contribution weight of each communication round of the participating client, m is the total number of participating clients in each round, T is the total number of iteration rounds, and ∑ represents the sum.

7. A contribution evaluation system based on value gain in federated learning, characterized in that: include: The first module is used to calculate the correlation between each client's local model and the global model in the tth round based on the contribution evaluation index of the correlation between the multi-dimensional computing models; The second module is used to calculate the parameter distance between the global model and the preheating model in the tth round based on the dynamic adjustment mechanism and perform normalization operation to obtain the contribution weight of each communication round of the participating clients; The third module is used to calculate the true contribution value of the client based on the correlation between each client's local model and the global model in the tth round and the contribution weight of each client participating in each communication round; The value-gain-based contribution evaluation system in federated learning is used to execute the steps in the value-gain-based contribution evaluation method in federated learning according to any one of claims 1 to 6.

Images