Homogenization fractional matching method, system, and electronic device
By determining the graph type and using the corresponding algorithm to assign matching scores to the edges, the problem of uneven resource allocation is solved, achieving balanced distribution of nodes and optimization of computing resources, thereby improving the system's fault tolerance and robustness.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SUZHOU MICRO ERA QUANTUM TECHNOLOGY CO LTD
- Filing Date
- 2026-03-06
- Publication Date
- 2026-06-05
AI Technical Summary
Existing maximum score matching and maximum weight score matching methods are prone to uneven resource allocation, with some nodes being overloaded or idle, and are sensitive to noise errors, making it difficult to achieve a balanced distribution of resources across nodes.
By determining whether the graph to be solved is unweighted or weighted, the maximum degree first algorithm or the maximum weight first algorithm is used to assign matching scores to the edges in the graph, and a uniform score matching scheme is designed to ensure the balanced distribution of resources on each node.
This achieves a balanced distribution of resources across nodes, improves fault tolerance and robustness, reduces computational resource consumption, avoids polarization of node capacity, and improves overall efficiency.
Smart Images
Figure CN122152523A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computer technology, and in particular to a uniform score matching method, a uniform score matching system, and an electronic device. Background Technology
[0002] Fractional matching is an important technique commonly used in graph theory and optimization problems. Its core idea is to allow edges to take numerical scores, not just integers, to achieve more flexible resource allocation than integer matching. Fractional matching has wide applications in both classical and quantum computing. For example, in classical computing, it can solve problems such as cloud computing resource allocation, network traffic optimization, and automated chip design. In quantum computing, it can solve the maximum energy problem in quantum many-body models.
[0003] In related technologies, maximum score matching or maximum weight score matching is commonly used to plan matching schemes. In scenarios where only the sum of matching results or the sum of weights needs to be maximized, global optimum may be achieved. However, maximum score matching or maximum weight score matching techniques often lead to uneven resource allocation, which may result in some nodes being overloaded while others are idle, making them prone to local failures and sensitive to noise errors. Summary of the Invention
[0004] The present invention is proposed to solve at least one of the above-mentioned problems. According to a first aspect of the present invention, a uniform score matching method is provided, the uniform score matching method comprising:
[0005] Obtain the graph to be solved and identify whether the graph is unweighted or weighted.
[0006] If the graph is unweighted, then the maximum degree first algorithm is used to assign matching scores to each edge in the graph to be solved.
[0007] If the graph is weighted, then the maximum weight priority algorithm is used to assign matching scores to each edge in the graph to be solved.
[0008] Based on the matching score and the graph to be solved, the uniform score matching result of the graph to be solved is obtained.
[0009] In one embodiment of the present invention, the step of assigning matching scores to each edge of the graph to be solved using the maximum degree first algorithm includes:
[0010] Initialize the capacity, degree, and validity of each vertex in the graph to be solved.
[0011] Select the vertex with the highest validity and denote it as the target vertex.
[0012] Determine whether there are any unassigned matching scores for each of the neighboring edges of the target vertex.
[0013] If there are adjacent edges whose matching scores are not assigned, then assign matching scores to the adjacent edges whose matching scores are not assigned.
[0014] Update the capacity, degree, and validity of the target vertex to zero.
[0015] Determine whether there are any neighboring vertices of the target vertex whose capacity is not zero.
[0016] If they exist, update the capacity, degree, and validity of the neighbors with non-zero capacity.
[0017] Determine whether the matching scores of each edge in the graph to be solved have been assigned.
[0018] If there are edges that have not been assigned matching scores, then return to the step of selecting the vertex with the highest validity.
[0019] If there are no edges without assigned matching scores, output the matching scores for each edge.
[0020] In one embodiment of the present invention, the step of initializing the capacity, degree, and validity of each vertex in the graph to be solved includes:
[0021] Set the capacity of the vertex to 1.
[0022] Assign the degree of a vertex the number of its neighboring edges.
[0023] The validity of a vertex is assigned the ratio of its degree to its capacity.
[0024] In one embodiment of the present invention, the step of assigning matching scores to adjacent edges that have not been assigned matching scores includes:
[0025] The following formula is used to match the adjacent edges without assigned scores. Assign a value to the matching score:
[0026]
[0027] in, , , This represents the k-th neighboring edge of the j-th vertex. Let the set of adjacent edges of the j-th vertex be denoted as . Let the set of neighboring vertices of the j-th vertex be denoted as . This represents the matching score of the k-th neighboring edge of the j-th vertex. This represents the validity of the j-th vertex.
[0028] In one embodiment of the present invention, the step of updating the capacity, degree, and validity of neighboring nodes whose capacity is not zero includes:
[0029] Update the capacity of neighboring nodes with non-zero capacity using the following formula:
[0030]
[0031] Update the degree of neighbors with non-zero capacity using the following formula:
[0032]
[0033] Update the validity of neighbors with non-zero capacity using the following formula:
[0034]
[0035] in, This represents the capacity of the k-th neighbor. This represents the degree of the k-th neighbor. This represents the validity of the k-th neighbor.
[0036] In one embodiment of the present invention, the step of assigning matching scores to each edge in the graph to be solved using the maximum weight priority algorithm includes:
[0037] Initialize the capacity, weight, and effective weight of each vertex in the graph to be solved.
[0038] Select the vertex with the largest effective weight and denote it as the target vertex.
[0039] Determine whether there are any unassigned matching scores for each of the neighboring edges of the target vertex.
[0040] If there are adjacent edges whose matching scores are not assigned, then assign matching scores to the adjacent edges whose matching scores are not assigned.
[0041] Update the capacity, weight, and effective weight of the target vertex to zero.
[0042] Determine whether there are any neighboring vertices of the target vertex whose capacity is not zero.
[0043] If they exist, update the capacity, weight, and effective weight of the neighboring nodes whose capacity is not zero.
[0044] Determine whether the matching scores of each edge in the graph to be solved have been assigned.
[0045] If there are edges that have not been assigned matching scores, then return to the step of selecting the vertex with the largest effective weight.
[0046] If there are no edges without assigned matching scores, output the matching scores for each edge.
[0047] In one embodiment of the present invention, the step of initializing the capacity, weight, and effective weight of each vertex in the graph to be solved includes:
[0048] Set the capacity of the vertex to 1.
[0049] The weight of a vertex is assigned as the sum of the weights of its neighboring edges.
[0050] The effective weight of a vertex is assigned as the ratio of the vertex's weight to its capacity.
[0051] In one embodiment of the present invention, the step of assigning matching scores to adjacent edges that have not been assigned matching scores includes:
[0052] The following formula is used to match the adjacent edges without assigned scores. Assign a value to the matching score:
[0053]
[0054] The step of updating the capacity, weight, and effective weight of neighboring nodes with non-zero capacity includes:
[0055] Update the capacity of neighboring nodes with non-zero capacity using the following formula:
[0056]
[0057] The weights of neighbors with non-zero capacity are updated using the following formula:
[0058]
[0059] The effective weights of neighbors with non-zero capacity are updated using the following formula:
[0060]
[0061] in, , , Let f represent the f-th neighboring edge of the i-th vertex. Let i represent the set of adjacent edges of the i-th vertex. Let i represent the set of neighboring vertices of the i-th vertex. Let f represent the matching score of the f-th neighboring edge of the i-th vertex. This represents the effective weight of the i-th vertex. Indicates adjacent edge The weight, This represents the capacity of the f-th neighbor. This represents the weight of the f-th neighbor. This represents the effective weight of the f-th neighbor.
[0062] According to a second aspect of the present invention, a uniform score matching system is provided, comprising: a preprocessing module, an unweighted graph processing module, a weighted graph processing module, and a postprocessing module.
[0063] The preprocessing module is used to obtain the graph to be solved and to identify whether the graph to be solved is an unweighted graph or a weighted graph;
[0064] The unweighted graph processing module is connected to the preprocessing module and is used to assign matching scores to each edge of the graph to be solved using the maximum degree first algorithm.
[0065] The weighted graph processing module is connected to the preprocessing module and is used to assign matching scores to each edge in the graph to be solved using the maximum weight priority algorithm.
[0066] The post-processing module is connected to the unweighted graph processing module and the weighted graph processing module, and is used to obtain the uniform score matching result of the graph to be solved based on the matching score and the graph to be solved.
[0067] According to a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory, wherein when the computer program is executed by the processor, it implements any of the above-described uniform score matching methods.
[0068] According to the homogenized score matching method, system, and electronic device provided in embodiments of the present invention, the homogenized score matching method first determines whether the graph to be solved is unweighted or weighted, and then uses the maximum degree first algorithm to assign values to each edge of the unweighted graph, or the maximum weight first algorithm to assign values to each edge of the weighted graph. This automatically designs a homogenized score matching scheme for both unweighted and weighted graphs, enabling a more balanced distribution of resources across nodes, avoiding polarization of node capacity, improving overall fault tolerance and robustness, and maximizing global benefits. Simultaneously, it reduces the consumption of computing resources and the computational burden. Attached Figure Description
[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0070] Figure 1 This is a flowchart illustrating a uniform score matching method provided in an embodiment of the present invention.
[0071] Figure 2 A flowchart illustrating a uniform score matching method provided in another embodiment of the present invention;
[0072] Figure 3 This is a schematic diagram of the diagram to be solved according to an embodiment of the present invention;
[0073] Figure 4 This is a schematic diagram of the graph to be solved after uniform fraction matching, according to an embodiment of the present invention.
[0074] Figure 5 This is a schematic diagram of the structure of a uniform fraction matching system provided in an embodiment of the present invention;
[0075] Figure 6 This is a hardware structure block diagram of a computer terminal for a uniform score matching method provided in an embodiment of the present invention. Detailed Implementation
[0076] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.
[0077] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described in order to avoid obscuring the invention.
[0078] It should be understood that the invention can be embodied in various forms and should not be construed as being limited to the embodiments set forth herein. Rather, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
[0079] To fully understand this invention, a detailed structure will be presented in the following description to illustrate the technical solution proposed by this invention. Optional embodiments of the invention are described in detail below; however, in addition to these detailed descriptions, the invention may have other embodiments.
[0080] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.
[0081] The first aspect of this invention provides a method for uniform score matching, such as... Figure 1 As shown, the uniform score matching method includes:
[0082] S1. Obtain the graph to be solved and identify whether it is an unweighted or weighted graph.
[0083] It's important to note that an unweighted graph means that each edge has no associated numerical weight; it merely represents the connection between vertices, and all edges are considered equivalent in terms of connectivity. A weighted graph, on the other hand, means that each edge has a weight, which represents the actual meaning of the connection between vertices.
[0084] S2, if it is an unweighted graph, then the maximum degree first algorithm is used to assign matching scores to each edge in the graph to be solved.
[0085] It should be noted that the maximum degree first algorithm is a greedy algorithm in graph theory. It can select the vertex with the highest degree or the highest effective degree from the currently unprocessed vertices for processing each time. After processing, the vertex is removed and the relevant data of the remaining vertices is updated. Then, vertices are selected again from the unprocessed vertices for processing, and the process is repeated until all vertices have been processed.
[0086] S3, if it is a weighted graph, then the maximum weight priority algorithm is used to assign matching scores to each edge in the graph to be solved.
[0087] It should be noted that the maximum weight priority algorithm is also a greedy algorithm in graph theory. It can prioritize processing the vertex with the largest weight or effective weight among the currently unprocessed vertices, remove the vertex after processing, update the relevant data of the remaining vertices, and select vertices from the unprocessed vertices again for processing, and repeat the process until all vertices have been processed.
[0088] S4. Based on the matching scores and the graph to be solved, obtain the uniform score matching result of the graph to be solved.
[0089] It should be noted that the obtained uniform score matching result is a uniform score matching scheme. This scheme, while ensuring that the sum of the matching scores is as close as possible to the optimal solution, tries to constrain the distribution range of the matching scores to make the distribution more balanced and avoid extreme values. In application scenarios such as the entanglement distribution in quantum systems, a small number of matching values are sacrificed to maximize global benefits. Therefore, the uniform score matching scheme obtained in this application can better solve problems such as cloud computing resource allocation, network traffic optimization, and chip design automation in the classical computing field, and can better solve the maximum energy problem of quantum many-body models in the quantum computing field.
[0090] The uniform score matching method of this invention first determines whether the graph to be solved is unweighted or weighted. Then, it uses the maximum degree first algorithm to assign values to the edges of the unweighted graph, or the maximum weight first algorithm to assign values to the edges of the weighted graph. This automatically designs a uniform score matching scheme for both unweighted and weighted graphs, resulting in a more balanced distribution of resources across nodes. This avoids polarization in node capacity, improves overall fault tolerance and robustness, and maximizes global benefits. Simultaneously, it reduces computational resource consumption and burden.
[0091] In some embodiments, such as Figure 2 As shown, the steps for assigning matching scores to each edge in the graph to be solved using the maximum degree first algorithm include:
[0092] S201, Initialize the capacity, degree, and validity of each vertex in the graph to be solved.
[0093] It should be noted that capacity can represent the total matching score allocated to neighboring edges, degree can represent the number of edges connected to the corresponding vertex, and validity can be determined based on the corresponding capacity and degree.
[0094] S202, select the vertex with the highest validity and denote it as the target vertex.
[0095] As an example, if there are multiple vertices with the highest validity, then one of them is randomly selected as the target vertex.
[0096] S203, determine whether there are any unassigned matching scores for each of the neighboring edges of the target vertex.
[0097] S204, if there are adjacent edges whose matching scores are not assigned, then assign matching scores to the adjacent edges whose matching scores are not assigned.
[0098] It should be noted that if there are no adjacent edges with unassigned matching scores, the subsequent step S205 will continue.
[0099] S205 updates the capacity, degree, and validity of the target vertex to zero.
[0100] It is worth mentioning that updating the capacity, degree, and validity of the target vertex to zero indicates that the resources of that vertex have been allocated and it will no longer participate in subsequent vertex selection and score matching, thus preventing the duplicate allocation of resources.
[0101] S206, determine whether there are any neighboring vertices of the target vertex with non-zero capacity.
[0102] S207, if it exists, then update the capacity, degree, and validity of the neighboring nodes whose capacity is not zero.
[0103] It is worth mentioning that by identifying neighboring vertices with non-zero capacity, we can determine which vertices still have unallocated resources. We can then update the relevant parameters of these neighboring vertices to facilitate subsequent resource allocation.
[0104] It should be noted that if there is no neighboring vertex of the target vertex with a capacity of non-zero, then proceed to the next step S208.
[0105] S208, determine whether the matching scores of each edge in the graph to be solved have been assigned.
[0106] S209: If there are no edges without a matching score, output the matching score corresponding to each edge.
[0107] S210: If there are edges that have not been assigned matching scores, return the step of selecting the vertex with the highest validity.
[0108] In this embodiment, by selecting the vertex with the highest validity as the target vertex strategy—that is, prioritizing the processing of vertices with denser connections—resources can be distributed more evenly across nodes. Simultaneously, by setting the relevant data of processed vertices to zero, redundant resource allocation is effectively avoided. Combined with dynamic updates to neighbor parameters, this ensures accurate subsequent vertex priority determination, thereby improving allocation efficiency.
[0109] In some embodiments, the steps of initializing the capacity, degree, and validity of each vertex in the graph to be solved include:
[0110] S2011, assign the capacity of the vertex to 1.
[0111] It should be noted that assigning a capacity of 1 to each vertex can prevent the uniformity of score matching from being affected by the initial resource differences.
[0112] S2012 assigns the degree of a vertex to the number of its neighboring edges.
[0113] S2013 assigns the validity of a vertex to the ratio of its degree to its capacity.
[0114] As an example, all vertices can be initialized. capacity ,Spend , effectiveness .in, Represents the set of all vertices. This represents the capacity of the a-th vertex. This represents the degree of the a-th vertex. This represents the validity of the a-th vertex. Let represent the set of adjacent edges of the a-th vertex.
[0115] In this embodiment, the capacity of vertices is reassigned in a unified manner to avoid uneven resource allocation before resource allocation. At the same time, the degree is defined by the number of adjacent edges and the validity is defined by the ratio of degree to capacity. This can accurately reflect the connection strength of vertices and also construct a priority index that takes into account both connectivity and resource quantity.
[0116] In some embodiments, the step of assigning matching scores to adjacent edges that have not been assigned matching scores includes:
[0117] The following formula is used to match the adjacent edges without assigned scores. Assign a value to the matching score:
[0118]
[0119] in, , , This represents the k-th neighboring edge of the j-th vertex. Let the set of adjacent edges of the j-th vertex be denoted as . Let the set of neighboring vertices of the j-th vertex be denoted as . This represents the matching score of the k-th neighboring edge of the j-th vertex. This represents the validity of the j-th vertex.
[0120] As an example, the matching scores of each neighboring edge are repeatedly assigned using the above formula until the matching scores of all neighboring edges of the j-th vertex have been determined.
[0121] In this embodiment, by using the above-described assignment method, all unassigned neighboring edges under the same target vertex can obtain equal scores, avoiding the distribution deviation of neighboring edges within the same vertex and ensuring the balance of local distribution.
[0122] In some embodiments, the step of updating the capacity, degree, and validity of neighboring nodes with non-zero capacity includes:
[0123] Update the capacity of neighboring nodes with non-zero capacity using the following formula:
[0124]
[0125] Specifically, by subtracting the capacity of the corresponding neighboring nodes This ensures that the total amount of resources allocated subsequently does not exceed the initial capacity, thereby avoiding over-allocation of resources.
[0126] Update the degree of neighbors with non-zero capacity using the following formula:
[0127]
[0128] Specifically, since the relevant data of the target vertex has been set to zero, it is equivalent to removing the target vertex from the graph to be solved. In subsequent calculations, the target vertex will no longer participate in the subsequent connection calculations, the number of adjacent edges of k will decrease by 1, and the degree should also be reduced by 1.
[0129] Update the validity of neighbors with non-zero capacity using the following formula:
[0130]
[0131] in, This represents the capacity of the k-th neighbor. This represents the degree of the k-th neighbor. This represents the validity of the k-th neighbor.
[0132] In this embodiment, the above-described update steps can prevent over-allocation of resources and ensure the rationality of resource allocation.
[0133] In some embodiments, such as Figure 2 As shown, the steps for assigning matching scores to each edge in the graph to be solved using the maximum weight first algorithm include:
[0134] S301, Initialize the capacity, weight, and effective weight of each vertex in the graph to be solved.
[0135] It should be noted that weights can represent the actual meaning of vertices, and effective weights can be determined based on the corresponding capacity and weights.
[0136] S302, select the vertex with the largest effective weight and denote it as the target vertex.
[0137] As an example, if there are multiple vertices with the highest effective weight, then one of them is randomly selected as the target vertex.
[0138] S303, determine whether there are any unassigned matching scores for each of the neighboring edges of the target vertex.
[0139] S304, if there are adjacent edges whose matching scores are not assigned, then assign matching scores to the adjacent edges whose matching scores are not assigned.
[0140] It should be noted that if there are no adjacent edges with unassigned matching scores, the subsequent step S305 will continue.
[0141] S305 updates the capacity, weight, and effective weight of the target vertex to zero.
[0142] It is worth mentioning that updating the capacity, weight, and effective weight of the target vertex to zero indicates that the resources of that vertex have been allocated and it will no longer participate in subsequent vertex selection and score matching, thus preventing the duplicate allocation of resources.
[0143] S306, determine whether there are any neighboring vertices of the target vertex with non-zero capacity.
[0144] S307, if it exists, then update the capacity, weight, and effective weight of the neighboring nodes whose capacity is not zero.
[0145] It is worth mentioning that by identifying neighboring vertices with non-zero capacity, we can determine which vertices still have unallocated resources. We can then update the relevant parameters of these neighboring vertices to facilitate subsequent resource allocation.
[0146] It should be noted that if there is no neighboring vertex of the target vertex with a capacity of non-zero, then proceed to the next step S308.
[0147] S308, determine whether the matching scores of each edge in the graph to be solved have been assigned.
[0148] S309: If there are no edges without a matching score, output the matching score corresponding to each edge.
[0149] S310, if there are edges that have not been assigned matching scores, return the step of selecting the vertex with the largest effective weight.
[0150] In this embodiment, by selecting the vertex with the highest effective weight as the target vertex, i.e., prioritizing the processing of vertices with more practical significance, the distribution of resources across nodes becomes more balanced. Simultaneously, by setting the relevant data of processed vertices to zero, redundant resource allocation is effectively avoided. Combined with dynamic updates to neighbor parameters, this ensures accurate subsequent vertex priority determination, thereby improving allocation efficiency.
[0151] In some embodiments, the step of initializing the capacity, weight, and effective weight of each vertex in the graph to be solved includes:
[0152] S3011, assigns the capacity of the vertex to 1.
[0153] It should be noted that assigning a capacity of 1 to each vertex can prevent the uniformity of score matching from being affected by the initial resource differences.
[0154] S3012 assigns the weight of a vertex to the sum of the weights of its neighboring edges.
[0155] S3013 assigns the effective weight of a vertex to the ratio of the vertex's weight to its capacity.
[0156] As an example, all vertices can be initialized. capacity Weight Effective weights .in, Represents the set of all vertices. This represents the capacity of the b-th vertex. This represents the weight of the b-th vertex. This represents the effective weight of the b-th vertex. This represents the weight of the edge between the a-th vertex and the g-th vertex.
[0157] In this embodiment, the capacity of vertices is reassigned in a unified manner to avoid uneven resource allocation before resource allocation. At the same time, the weight of a vertex is defined by the weight set of the edges, and the effective weight is defined by the ratio of the weight to the capacity. This can accurately reflect the connection strength of the vertex and also construct a priority index that takes into account both connectivity and resource quantity.
[0158] In some embodiments, the step of assigning matching scores to adjacent edges that have not been assigned matching scores includes:
[0159] The following formula is used to match the adjacent edges without assigned scores. Assign a value to the matching score:
[0160]
[0161] in, , , Let f represent the f-th neighboring edge of the i-th vertex. Let i represent the set of adjacent edges of the i-th vertex. Let i represent the set of neighboring vertices of the i-th vertex. Let f represent the matching score of the f-th neighboring edge of the i-th vertex. This represents the effective weight of the i-th vertex. Indicates adjacent edge The weight.
[0162] As an example, the matching scores of each neighboring edge are repeatedly assigned using the above formula until the matching scores of all neighboring edges of the i-th vertex have been determined.
[0163] In this embodiment, the above-mentioned assignment method calculates the matching score by the ratio of the weight of the neighboring edge itself to the effective weight of the vertex. This allows the neighboring edge with higher weight to obtain a higher matching score, making the allocation result fit the intrinsic importance difference of the edge and ensuring the difference in local allocation.
[0164] In some embodiments, the step of updating the capacity, weight, and effective weight of neighbors with non-zero capacity includes:
[0165] Update the capacity of neighboring nodes with non-zero capacity using the following formula:
[0166]
[0167] Specifically, by subtracting the capacity of the corresponding neighboring nodes This ensures that the total amount of resources allocated subsequently does not exceed the initial capacity, thereby avoiding over-allocation of resources.
[0168] The weights of neighbors with non-zero capacity are updated using the following formula:
[0169]
[0170] The effective weights of neighbors with non-zero capacity are updated using the following formula:
[0171]
[0172] in, This represents the capacity of the f-th neighbor. This represents the weight of the f-th neighbor. This represents the effective weight of the f-th neighbor.
[0173] In this embodiment, the above-described update steps can prevent over-allocation of resources and ensure the rationality of resource allocation.
[0174] Next, the present invention will... Figure 3 The diagram to be solved is shown. Taking an example, the specific process of the uniformized score matching method of the present invention will be explained:
[0175] A1, Connect the graph to be solved. It also identified the graph as an unweighted graph.
[0176] A2 assigns matching scores to each edge in the graph to be solved using the maximum degree first algorithm.
[0177] Specifically, it includes:
[0178] I. Initialize the capacity, degree, and validity of all vertices:
[0179]
[0180]
[0181]
[0182]
[0183]
[0184] II. Select the vertex with the highest validity, i.e., vertex 0, as the target vertex.
[0185] III. Adjacent edges of vertex 0 , , , The matching scores are all undetermined, so they are assigned the value of validity. The reciprocal of, that is .
[0186] IV. Update the capacity, degree, and validity of vertex 0 to 0, i.e. .
[0187] V. If the capacities of vertex 0's neighbors 1, 2, 3, and 4 are non-zero and have not been updated in this round, update them:
[0188]
[0189]
[0190]
[0191]
[0192] VI. Select the vertex with the highest validity. If there are multiple vertices with the highest validity, randomly select one of them, for example, select vertex 1.
[0193] VII. Adjacent edges of vertex 1 , The matching score is not determined; the value is assigned as the validity score. The reciprocal of, that is .
[0194] VIII. Update the capacity, degree, and validity of vertex 1 to 0, i.e. .
[0195] IX. Vertex 1 has neighbors vertices 2 and 4 whose capacities are non-zero and have not been updated in this round. Update them as follows:
[0196]
[0197]
[0198] X. Select the vertex with the highest validity. If there are multiple vertices with the highest validity, select one of them randomly, for example, select vertex 3.
[0199] VII. Adjacent edges of vertex 3 , The matching score is not determined; the value is assigned as the validity score. The reciprocal of, that is .
[0200] VIII. All edges in the graph are assigned matching scores, and the maximum degree first algorithm ends.
[0201] A3, Figure to be solved Since it is an unweighted graph, no operations related to weighted graphs will be performed.
[0202] A4, connect the graph after processing in step A2 above. The resulting uniform fraction matching scheme is as follows: Figure 4 As shown, the numbers on each edge represent the matching score of that edge.
[0203] In addition, the present invention also provides a uniformized score matching system, such as Figure 5 As shown, the uniform score matching system includes: a preprocessing module 10, an unweighted graph processing module 20, a weighted graph processing module 30, and a postprocessing module 40.
[0204] The preprocessing module 10 is used to obtain the graph to be solved and identify whether the graph to be solved is an unweighted graph or a weighted graph;
[0205] The unweighted graph processing module 20 is connected to the preprocessing module 10 and is used to assign matching scores to each edge of the graph to be solved using the maximum degree first algorithm.
[0206] The weighted graph processing module 30 is connected to the preprocessing module 10 and is used to assign matching scores to each edge in the graph to be solved using the maximum weight priority algorithm.
[0207] The post-processing module 40 is connected to the unweighted graph processing module 20 and the weighted graph processing module 30, and is used to obtain the uniform score matching result of the graph to be solved based on the matching score and the graph to be solved.
[0208] Other specific embodiments of the homogenized score matching system of the present invention can be found in the specific embodiments of the homogenized score matching method of the above embodiments of the present invention.
[0209] The uniform score matching system of this invention first determines whether the graph to be solved is unweighted or weighted by the preprocessing module 10. Then, it uses the maximum degree first algorithm in the unweighted graph processing module 20 to assign values to the edges of the graph to be solved if it is unweighted, or the maximum weight first algorithm in the weighted graph processing module 30 to assign values to the edges of the graph to be solved if it is weighted. This automatically designs a uniform score matching scheme for the unweighted or weighted graph, making the distribution of resources more balanced across nodes, avoiding polarization of node capacity, improving overall fault tolerance and robustness, and maximizing global benefits. At the same time, it reduces the consumption of computing resources and the computational burden.
[0210] The following detailed explanation uses a computer terminal as an example. Figure 6 This is a hardware structure block diagram of a computer terminal for a uniform score matching method provided in an embodiment of the present invention. Figure 6As shown, a computer terminal may include one or more ( Figure 6 Only one is shown in the diagram. A processor 601 (processor 601 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 602 for storing data are also shown. Optionally, the computer terminal may further include a transmission device 603 for communication functions and an input / output device 604. Those skilled in the art will understand that... Figure 6 The structure shown is for illustrative purposes only and does not limit the structure of the computer terminal described above. For example, the computer terminal may also include components that are more complex than those described above. Figure 6 The more or fewer components shown, or having the same Figure 6 The different configurations shown.
[0211] The memory 602 can be used to store software programs and modules of application software, such as the program instructions / modules corresponding to the uniform score matching method in this embodiment. The processor 601 executes various functional applications and data processing by running the software programs and modules stored in the memory 602, thereby implementing the above-described method. The memory 602 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 602 may further include memory remotely located relative to the processor 601, and these remote memories can be connected to a computer terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0212] The transmission device 603 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal. In one example, the transmission device 603 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 603 may be a Radio Frequency (RF) module for wireless communication with the Internet. Embodiments of this application also provide a computer-readable storage medium storing a computer program for electronic data interchange, which causes a computer to perform some or all of the steps of any of the methods described in the above method embodiments, wherein the computer includes an electronic device.
[0213] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may include an electronic device.
[0214] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of this application. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of this application. All such changes and modifications are intended to be included within the scope of this application as claimed in the appended claims.
[0215] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0216] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.
[0217] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.
[0218] Similarly, it should be understood that, in order to streamline this application and aid in understanding one or more of the various inventive aspects, features of this application may sometimes be grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of this application. However, this approach should not be construed as reflecting an intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with features fewer than all features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of this application.
[0219] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or elements of any method or apparatus so disclosed may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.
[0220] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are intended to be within the scope of this application and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.
[0221] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules according to the embodiments of this application. This application can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
[0222] It should be noted that the above embodiments are illustrative of this application and not limiting of it, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware.
[0223] The above description is merely a specific embodiment or illustration of the embodiments of this application. The scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. The scope of protection of this application shall be determined by the scope of the claims.
Claims
1. A method for uniformly matching fractions, characterized in that, The method includes: Obtain the graph to be solved and identify whether the graph to be solved is an unweighted graph or a weighted graph; If the graph is unweighted, then the maximum degree first algorithm is used to assign matching scores to each edge in the graph to be solved. If the graph is weighted, then the maximum weight first algorithm is used to assign matching scores to each edge in the graph to be solved. Based on the matching score and the graph to be solved, the uniform score matching result of the graph to be solved is obtained.
2. The homogenized fraction matching method according to claim 1, characterized in that, The step of assigning matching scores to each edge in the graph to be solved using the maximum degree first algorithm includes: Initialize the capacity, degree, and validity of each vertex in the graph to be solved; Select the vertex with the highest validity and denote it as the target vertex; Determine whether any of the neighboring edges of the target vertex have unassigned matching scores; If there are adjacent edges whose matching scores are not assigned, then assign matching scores to the adjacent edges whose matching scores are not assigned. Update the capacity, degree, and validity of the target vertex to zero; Determine whether there are any neighboring vertices of the target vertex with a non-zero capacity; If they exist, update the capacity, degree, and validity of the neighboring nodes whose capacity is not zero; Determine whether the matching scores of each edge in the graph to be solved have been assigned; If there are edges that have not been assigned matching scores, then return to the step of selecting the vertex with the highest validity. If there are no edges without assigned matching scores, output the matching scores for each edge.
3. The homogenized fraction matching method according to claim 2, characterized in that, The steps of initializing the capacity, degree, and validity of each vertex in the graph to be solved include: Set the capacity of each vertex to 1; Assign the degree of a vertex the number of its neighboring edges; The validity of a vertex is assigned the ratio of its degree to its capacity.
4. The homogenized fraction matching method according to claim 2, characterized in that, The step of assigning matching scores to adjacent edges that have not been assigned matching scores includes: The following formula is used to match the adjacent edges without assigned scores. Assign a value to the matching score: ； in, , , This represents the k-th neighboring edge of the j-th vertex. Let the set of adjacent edges of the j-th vertex be denoted as . Let the set of neighboring vertices of the j-th vertex be denoted as . This represents the matching score of the k-th neighboring edge of the j-th vertex. This represents the validity of the j-th vertex.
5. The homogenized fraction matching method according to claim 4, characterized in that, The step of updating the capacity, degree, and validity of neighboring nodes with non-zero capacity includes: Update the capacity of neighboring nodes with non-zero capacity using the following formula: ； Update the degree of neighbors with non-zero capacity using the following formula: ； Update the validity of neighbors with non-zero capacity using the following formula: ； in, This represents the capacity of the k-th neighbor. This represents the degree of the k-th neighbor. This represents the validity of the k-th neighbor.
6. The homogenized fraction matching method according to claim 1, characterized in that, The step of assigning matching scores to each edge in the graph to be solved using the maximum weight first algorithm includes: Initialize the capacity, weight, and effective weight of each vertex in the graph to be solved; Select the vertex with the largest effective weight and denote it as the target vertex; Determine whether any of the neighboring edges of the target vertex have unassigned matching scores; If there are adjacent edges whose matching scores are not assigned, then assign matching scores to the adjacent edges whose matching scores are not assigned. Update the capacity, weight, and effective weight of the target vertex to zero; Determine whether there are any neighboring vertices of the target vertex with a non-zero capacity; If they exist, update the capacity, weight, and effective weight of the neighboring nodes whose capacity is not zero; Determine whether the matching scores of each edge in the graph to be solved have been assigned; If there are edges that have not been assigned matching scores, return to the step of selecting the vertex with the largest effective weight; If there are no edges without assigned matching scores, output the matching scores for each edge.
7. The homogenized fraction matching method according to claim 6, characterized in that, The steps of initializing the capacity, weight, and effective weight of each vertex in the graph to be solved include: Set the capacity of each vertex to 1; The weight of a vertex is assigned as the sum of the weights of its neighboring edges; The effective weight of a vertex is assigned as the ratio of the vertex's weight to its capacity.
8. The homogenized fraction matching method according to claim 6, characterized in that, The step of assigning matching scores to adjacent edges that have not been assigned matching scores includes: The following formula is used to match the adjacent edges without assigned scores. Assign a value to the matching score: ； The step of updating the capacity, weight, and effective weight of neighboring nodes with non-zero capacity includes: Update the capacity of neighboring nodes with non-zero capacity using the following formula: ； The weights of neighbors with non-zero capacity are updated using the following formula: ； The effective weights of neighbors with non-zero capacity are updated using the following formula: ； in, , , Let f represent the f-th neighboring edge of the i-th vertex. Let i represent the set of adjacent edges of the i-th vertex. Let i represent the set of neighboring vertices of the i-th vertex. Let f represent the matching score of the f-th neighboring edge of the i-th vertex. This represents the effective weight of the i-th vertex. Indicates adjacent edge The weight, This represents the capacity of the f-th neighbor. This represents the weight of the f-th neighbor. This represents the effective weight of the f-th neighbor.
9. A uniform fraction matching system, characterized in that, The system includes: The preprocessing module is used to obtain the graph to be solved and identify whether the graph to be solved is an unweighted graph or a weighted graph; An unweighted graph processing module, connected to the preprocessing module, is used to assign matching scores to each edge in the graph to be solved using a maximum degree first algorithm. A weighted graph processing module, connected to the preprocessing module, is used to assign matching scores to each edge in the graph to be solved using the maximum weight first algorithm. The post-processing module, connected to the unweighted graph processing module and the weighted graph processing module, is used to obtain the uniformized score matching result of the graph to be solved based on the matching score and the graph to be solved.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, When the computer program is executed by the processor, it implements the uniform score matching method according to any one of claims 1-8.