A resource allocation method, apparatus, device, medium, and product

Abstract

This application provides a resource allocation method, apparatus, device, medium, and product, which can be applied to the fields of distributed technology and fintech. The method includes: determining an initial resource lending scheme for each lender in the set of lenders and borrowers based on a set of lenders and borrowers; defining the initial resource lending scheme of each lender as its current resource lending scheme, and iteratively executing the following preset steps until a preset loop termination condition is met: calling a computing node to update the amount of resources lent to each target borrower in the current resource lending scheme of each lender based on a preset optimization objective, reducing the target resource amount, and obtaining a new resource lending scheme for each lender; and, if the new resource lending scheme of each lender meets a preset update condition, defining the new resource lending scheme of each lender as its new current resource lending scheme.

Description

Technical Field

[0001] This application relates to the fields of distributed technology and financial technology, specifically to a resource allocation method, apparatus, device, medium, and product. Background Technology

[0002] Currently, for multiple lending institutions with resources and multiple borrowing institutions that need to borrow resources, the lending plan usually needs to be determined manually. For example, if a borrowing institution needs to borrow resources from one or more lending institutions, or a lending institution needs to lend resources to one or more borrowing institutions, the specific allocation of resources also needs to be determined manually.

[0003] As the number of institutions increases, manually determining resource lending schemes becomes less efficient. Summary of the Invention

[0004] In view of the above problems, this application provides a resource allocation method, apparatus, equipment, medium and product for improving the efficiency of determining resource lending schemes.

[0005] According to a first aspect of this application, a resource allocation method is provided, applied to a control node; the control node is used to control a computing node; the method includes: determining a set of lending institutions and a set of borrowing institutions; any lending institution lending resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, determining an initial resource lending scheme for each lending institution in the set of lending institutions; the resource lending scheme includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending schemes of each lending institution satisfy preset constraints; determining the initial resource lending schemes of each lending institution as the current resource lending schemes of each lending institution, and cyclically executing the following preset steps until a preset cycle stop is satisfied. Termination Conditions: First, identify the target lending institution and determine the target resource amount that the target lending institution will lend to any target borrowing institution. Second, invoke the computing node to update the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution based on a preset optimization target, reducing the target resource amount to obtain a new resource lending plan for each lending institution. The new resource lending plan of each lending institution satisfies the preset constraint condition. If the new resource lending plan of each lending institution satisfies the preset update condition, the new resource lending plan of each lending institution is determined as the new current resource lending plan of each lending institution. The preset update condition includes: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

[0006] According to an embodiment of this application, the preset step further includes at least one of the following: if the new resource lending plan of each lending institution does not meet the preset update conditions, the current resource lending plan of each lending institution before the update is re-determined as the current resource lending plan of each lending institution; if no new resource lending plan of each lending institution is determined within a preset time threshold, the current resource lending plan of each lending institution before the update is re-determined as the current resource lending plan of each lending institution.

[0007] According to embodiments of this application, the preset optimization objective includes at least one of the following: maximizing the resource reduction of the target resource quantity; minimizing the absolute value of resource changes in other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution; minimizing the number of resource changes among other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution; maximizing the comprehensive change value of the resource quantity; wherein the comprehensive change value of the resource quantity is positively correlated with the resource reduction of the target resource quantity, the comprehensive change value of the resource quantity is negatively correlated with the absolute value of resource changes in other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution, and the comprehensive change value of the resource quantity is negatively correlated with the number of resource changes among other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution.

[0008] According to an embodiment of this application, the control node is used to control multiple computing nodes; the step of determining the target lending institution and determining the target resource amount lent by the target lending institution to any target borrowing institution includes: determining the correspondence between different groups of target lending institutions and target resource amounts; the step of calling the computing node to update the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution based on a preset optimization target, reducing the target resource amount, and obtaining a new resource lending plan for each lending institution, wherein the new resource lending plan of each lending institution satisfies the preset constraint conditions. This includes: Parallel invocation of different computing nodes to perform the following operations on the correspondence between different groups of target lending institutions and target resource amounts, resulting in a corresponding set of new resource lending schemes: Based on a preset optimization objective, updating the resource amount lent to each target borrowing institution in the current resource lending scheme of each lending institution, reducing the target resource amount, and obtaining a new resource lending scheme for each lending institution as a corresponding set of new resource lending schemes. The new resource lending schemes of each lending institution satisfy the preset constraint conditions; the preset update conditions also include: a preset attribute that is superior to any other set of new resource lending schemes.

[0009] According to embodiments of this application, determining the target lending institution includes any of the following: identifying any lending institution in the current resource lending scheme whose number of target borrowing institutions exceeds a preset threshold as the target lending institution; identifying the lending institution in the current resource lending scheme with the largest number of target borrowing institutions exceeding the preset threshold as the target lending institution; and determining the target lending institution based on the number of target borrowing institutions among the lending institutions in the current resource lending scheme whose number of target borrowing institutions exceeds the preset threshold; wherein the probability of a lending institution being identified as a target lending institution is positively correlated with the number of target borrowing institutions.

[0010] According to an embodiment of this application, determining the target resource amount lent by the target lending institution to any target borrowing institution includes any one of the following: determining the minimum resource amount lent by the target lending institution to the target borrowing institution as the target resource amount; determining the target resource amount based on the resource amount lent by the target lending institution to the target borrowing institution; wherein the probability of the resource amount being determined as the target resource amount is negatively correlated with the size of the resource amount.

[0011] According to an embodiment of this application, the preset loop stopping condition includes at least one of the following: the number of loops is greater than a preset number threshold; the number of target borrowing institutions in the current resource lending scheme of each lending institution is less than or equal to a preset institution number threshold; and the number of lending institutions borrowing resources by each borrowing institution satisfies a preset sparsity constraint.

[0012] According to embodiments of this application, the preset constraints include at least one of the following: the resource amount of each lending institution is equal to the total resource amount lent to each target borrowing institution in the corresponding resource lending scheme; the resource demand of each borrowing institution is equal to the total resource amount lent by each lending institution; the resource amount lent by each lending institution to any borrowing institution is greater than a preset lower limit threshold for resource amount; and for each lending institution, in the corresponding resource lending scheme, the proportion of the total resource interest of each target borrowing institution in the resource amount of the targeted lending institution is within a preset proportion range.

[0013] The second aspect of this application provides another resource allocation method, comprising: determining a set of lending institutions and a set of borrowing institutions; any lending institution lending resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, determining an initial resource lending scheme for each lending institution in the set of lending institutions; the resource lending scheme includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending schemes of each lending institution satisfy preset constraints; determining the initial resource lending schemes of each lending institution as the current resource lending schemes of each lending institution, and cyclically executing the following preset steps until a preset cycle stop condition is met: determining target lending institutions. The system constructs a new resource lending plan for each lending institution and determines the target amount of resources to be lent by the target lending institution to any target borrowing institution. Based on a preset optimization target, the system updates the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution, reduces the target resource amount, and obtains a new resource lending plan for each lending institution. The new resource lending plan of each lending institution satisfies the preset constraint condition. If the new resource lending plan of each lending institution satisfies the preset update condition, the new resource lending plan of each lending institution is determined as the new current resource lending plan of each lending institution. The preset update condition includes: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

[0014] A third aspect of this application provides a resource allocation apparatus applied to a control node; the control node is used to control a computing node; the apparatus includes: a first initialization module, configured to determine a set of lending institutions and a set of borrowing institutions; any lending institution is used to lend resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, an initial resource lending scheme is determined for each lending institution in the set of lending institutions; the resource lending scheme includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending schemes of each lending institution satisfy preset constraints; a first looping module, configured to determine the initial resource lending schemes of each lending institution as the current resource lending schemes of each lending institution, and loop through the following preset steps until... The following preset loop stopping conditions are met: A target lending institution is identified, and the target resource amount lent by that institution to any target borrowing institution is determined. The computing node is invoked, and based on a preset optimization target, the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution is updated, reducing the target resource amount to obtain a new resource lending plan for each institution. The new resource lending plan of each institution satisfies the preset constraint condition. If the new resource lending plan of each institution satisfies the preset update condition, the new resource lending plan of each institution is determined as the new current resource lending plan for each institution. The preset update condition includes: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

[0015] A fourth aspect of this application provides another resource allocation apparatus, comprising: an application to a control node; the control node being used to control a computing node; the apparatus comprising: a first initialization module, configured to determine a set of lending institutions and a set of borrowing institutions; any lending institution is used to lend resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, determining an initial resource lending scheme for each lending institution in the set of lending institutions; the resource lending scheme includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending schemes for each lending institution satisfy preset constraints; a first looping module, configured to determine the initial resource lending schemes of each lending institution as the current resource lending schemes of each lending institution, and cyclically execute the following preset steps. The process continues until a preset loop stopping condition is met: A target lending institution is identified, and the target resource amount lent by that institution to any target borrowing institution is determined. The computing node is invoked, and based on a preset optimization target, the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution is updated, reducing the target resource amount to obtain a new resource lending plan for each institution. The new resource lending plan of each institution satisfies the preset constraint condition. If the new resource lending plan of each institution satisfies a preset update condition, the new resource lending plan of each institution is determined as the new current resource lending plan for each institution. The preset update condition includes: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

[0016] A fifth aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0017] A sixth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0018] A seventh aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method. Attached Figure Description

[0019] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0020] Figure 1This illustration schematically depicts an application scenario of a resource allocation method according to an embodiment of this application.

[0021] Figure 2 A flowchart illustrating a resource allocation method according to an embodiment of this application is shown schematically;

[0022] Figure 3 A flowchart illustrating another resource allocation method according to an embodiment of this application is shown schematically;

[0023] Figure 4 This schematic diagram illustrates a structural block diagram of a resource allocation device according to an embodiment of the present application;

[0024] Figure 5 This schematic diagram illustrates a structural block diagram of another resource allocation device according to an embodiment of the present application;

[0025] Figure 6 A block diagram schematically illustrates an electronic device suitable for implementing a resource allocation method according to an embodiment of this application. Detailed Implementation

[0026] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0027] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0028] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0029] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0030] Currently, for multiple lending institutions with resources and multiple borrowing institutions needing to borrow resources, the lending scheme typically needs to be determined manually. For example, a borrowing institution may need to borrow resources from one or more lending institutions, or a lending institution may need to lend resources to one or more borrowing institutions; the specific allocation of resources also needs to be determined manually. As the number of institutions increases, the efficiency of manually determining the lending scheme becomes low. The resources involved are not limited; they can be funds, virtual resources (e.g., cloud resources, storage resources, computing power resources, etc.), or physical resources (e.g., commodity resources, goods resources, warehousing resources, etc.).

[0031] To address the aforementioned technical problems and improve the efficiency of determining resource lending schemes, embodiments of this application provide a resource allocation method. In this method, resource lending schemes can be solved using computer equipment, thereby improving the efficiency of determining resource lending schemes.

[0032] Furthermore, this method can be based on an iterative improvement process to progressively optimize the problem of determining resource lending schemes, thereby reducing the problem size and difficulty of solving the problem and improving the efficiency of determining resource lending schemes. Specifically, when determining the resource lending schemes for each lending institution, it is usually necessary to set constraints. Among these constraints, constraints can be set to control the borrowing institution to borrow resources from as few lending institutions as possible, and to control the number of resource lending relationships to be as small as possible, so that the number of lending institutions for which each borrowing institution borrows resources is less than a preset threshold for the number of lending institutions.

[0033] To address this constraint, an iterative improvement process can be used. Specifically, initially, the constraint can be disregarded, and an initial version of the resource lending scheme for each lending institution can be determined. Then, through iterative improvement, the number of borrowing institutions in each lending institution's resource lending scheme can be gradually reduced, updating the resource lending scheme for each institution until the constraint is satisfied. Each iteration can target the reduction of a certain number of resource lending relationships. For example, reducing at least one resource lending relationship, or reducing one borrowing institution from any lending institution's resource lending scheme, can be the objective of a single iteration, updating the resource lending scheme accordingly. This reduces the difficulty of solving the problem and improves the efficiency of determining the resource lending scheme.

[0034] The above method can decompose a constraint into multiple sub-problems that reduce the number of resource lending relationships through multiple iterations, thereby reducing the difficulty of solving the problem and improving the efficiency of determining resource lending schemes.

[0035] It should be noted that the resource allocation method and apparatus provided in the embodiments of this application can be applied to the fields of distributed technology and fintech, as well as any field other than fintech. For example, in the field of cloud computing technology, the resource allocation method provided in the embodiments of this application can be used to determine the resource lending scheme for the borrowing and allocation of cloud resources. The application fields of the resource allocation method and apparatus provided in the embodiments of this application are not limited.

[0036] In the technical solution of this application, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0037] In scenarios involving automated decision-making using personal information, the methods, devices, and systems provided in this application all offer users corresponding entry points for choosing to agree to or reject the automated decision-making results. If the user chooses to reject, the process proceeds to the expert decision-making stage. Here, "automated decision-making" refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, and credit status through computer programs, and then making a decision. Here, "expert decision-making" refers to the activity of making decisions by personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0038] Figure 1 The illustration depicts an application scenario of a resource allocation method according to an embodiment of this application. For example... Figure 1 As shown, the application scenario 100 according to this embodiment may include: a first computing node 101, a second computing node 102, a third computing node 103, and a control node 104.

[0039] The first computing node 101, the second computing node 102, the third computing node 103, and the control node 104 can be different nodes in a distributed system. Specifically, they can be logically different nodes, nodes deployed on different devices, different server nodes, etc. The control node 104 can be used to control the first computing node 101, the second computing node 102, and the third computing node 103. It should be noted that the resource allocation method provided in this application embodiment can generally be executed by the control node 104. Correspondingly, the resource allocation device provided in this application embodiment can generally be set in the control node 104. It should be understood that... Figure 1 The number of distributed nodes shown is merely illustrative. Depending on implementation requirements, there can be any number of distributed nodes.

[0040] Figure 2 A flowchart illustrating a resource allocation method according to an embodiment of this application is shown schematically. Figure 2 As shown, the resource allocation method provided in this application embodiment may include operations S210 to S230. The embodiments of this application do not limit the specific executing entity of the resource allocation method; it can be executed by any electronic device or any software application. Optionally, a resource allocation method can be executed by a control node. The control node can be used to control computing nodes, and the control node and the controlled computing nodes can be different nodes in the same distributed system.

[0041] In operation S210, the set of lending institutions and the set of borrowing institutions are determined; any lending institution is used to lend resources to one or more borrowing institutions.

[0042] In operation S220, based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as the target of resource lending, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints.

[0043] In operation S230, the initial resource lending plan of each lending institution is determined as the current resource lending plan of each lending institution. The following preset steps are executed cyclically until a preset loop stop condition is met: 1. Determine the target lending institution and the target resource amount to be lent by the target lending institution to any target borrowing institution; 2. Call the computing node to update the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution based on a preset optimization target, reducing the target resource amount to obtain a new resource lending plan for each lending institution, and ensuring that the new resource lending plan of each lending institution meets preset constraints; 3. If the new resource lending plan of each lending institution meets preset update conditions, the new resource lending plan of each lending institution is determined as the new current resource lending plan of each lending institution; the preset update conditions include: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

[0044] The above method first determines the initial resource lending scheme through the control node, and then gradually reduces the number of resource lending relationships by iteratively optimizing the resource lending scheme. This can optimize the resource lending scheme, reduce the difficulty and computational load of determining the resource lending scheme, and improve the efficiency of determining the resource lending scheme.

[0045] The embodiments of this application do not limit the lending institution and the borrowing institution. Optionally, the lending institution can be an institution with resources that can lend resources; the borrowing institution can be an institution that needs to borrow resources and can borrow resources from other institutions. The embodiments of this application also do not limit the form of the lent resources. Optionally, the resources can be funds, virtual resources, or physical resources, etc. In a specific example, the lending institution can lend funds to the borrowing institution, and the lending institution can also lend virtual resources to the borrowing institution, such as computing power resources, storage resources, cloud resources, etc. Specifically, the lending institution can rent out its own computing power equipment, while the borrowing institution can borrow the computing power equipment provided by the lending institution for computation, specifically image processing, video processing, model training, and other operations.

[0046] The embodiments of this application do not limit the specific methods for determining the set of lending institutions and the set of borrowing institutions. Optionally, it may be that lending institutions that need to lend resources are added to the set of lending institutions, and borrowing institutions that need to borrow resources are added to the set of borrowing institutions; wherein, corresponding constraints may be additionally set, for example, the total amount of resources that can be lent in the determined set of lending institutions is equal to or greater than the total resource demand of the set of borrowing institutions, so that the resource amount of the set of lending institutions can meet the resource demand of the set of borrowing institutions.

[0047] The embodiments of this application do not limit the specific form of the resource lending scheme. Optionally, corresponding to a single lending institution or a subset of single lending institutions, the resource lending scheme may include: one or more target borrowing institutions as the target of resource lending, and the amount of resources lent to each target borrowing institution.

[0048] For ease of description, the relationship of a lending institution lending resources to a borrowing institution is referred to as a resource lending relationship. Accordingly, a lending institution can have resource lending relationships with different target borrowing institutions in a corresponding resource lending scheme. Optionally, a resource lending relationship can include a lending institution, a borrowing institution, and the amount of resources lent by the lending institution to the borrowing institution. Through iterative optimization in operation S230, the number of target borrowing institutions corresponding to a lending institution can be gradually reduced, which means reducing the corresponding resource lending relationships, thereby gradually reducing the total number of resource lending relationships.

[0049] In operation S220, for both the lending institution set and the borrowing institution set, an initial version of the resource lending scheme for each lending institution can be preliminarily determined. This initial version serves as the starting point for iterative optimization, satisfying basic constraints. This allows for a gradual reduction in the number of resource lending relationships through iterative optimization, thereby optimizing the resource lending scheme and improving the efficiency of determining the scheme. Compared to directly determining a single resource lending scheme that satisfies all constraints and minimizes resource lending relationships for a large number of lending and borrowing institutions, iterative optimization reduces the difficulty and computational load of determining the resource lending scheme, thus improving efficiency.

[0050] The embodiments of this application are not limited to the preset constraints in S220. Specific basic constraints can be set to reduce the difficulty and computational load in determining the initial resource lending scheme. The preset constraints can be set according to actual business needs.

[0051] For ease of understanding, the embodiments of this application provide specific examples of some preset constraints: (1) Since the interest rates offered by different lending institutions may be different, in order to improve fairness, the total interest rate obtained by different lending institutions can be balanced as much as possible, that is, the ratio of the final income to the amount of resources among different lending institutions, and should be kept as stable as possible within a specified range. (2) In order to reduce the difficulty of determining the resource lending scheme and improve the efficiency of determining the resource lending scheme, the lower limit of the amount of resources lent for each transaction can be limited, or the amount of resources lent for each transaction can be limited to a positive integer multiple of the lower limit of the amount of resources. (3) In the final resource lending scheme of each lending institution, the resource demand of each borrowing institution should be the same as the total amount of resources borrowed by the borrowing institution in the resource lending scheme of each lending institution, so as to meet the resource demand of the borrowing institution. In addition, the amount of resources lent by each lending institution can be the same as the amount of resources reserved by the lending institution itself or the amount of resources used for lending. (4) In addition, some other business constraints can be set, such as specifying that a certain lending institution cannot lend resources to some borrowing institutions, or specifying that a certain borrowing institution cannot borrow resources from some lending institutions, etc.

[0052] Optionally, the preset constraints include at least one of the following: (1) the resource amount of each lending institution is equal to the total resource amount lent to each target borrowing institution in the corresponding resource lending scheme; (2) the resource demand of each borrowing institution is equal to the total resource amount lent by each lending institution; (3) the resource amount lent by each lending institution to any borrowing institution is greater than the preset lower limit threshold of the resource amount; (4) for each lending institution, in the corresponding resource lending scheme, the proportion of the total resource interest of each target borrowing institution in the resource amount of the targeted lending institution is within the preset proportion range. This embodiment can conveniently set different preset constraints based on specific preset constraints to adapt to different business needs and improve the adaptability of the determined resource lending scheme to business needs.

[0053] It should be noted that the preset constraints may not include constraints on the number of resource lending relationships, which can easily reduce the difficulty and computational load of determining the initial resource lending scheme. The number of resource lending relationships can be reduced by operating S230.

[0054] The resource requirement of each borrowing institution can be equal to the total amount of resources borrowed from each lending institution, which is the total amount of resources borrowed by the target borrowing institution in the resource lending plan of each lending institution.

[0055] It is understood that the specific embodiments of the above-mentioned preset constraints are merely illustrative examples, and other preset constraints can be set according to actual needs.

[0056] In a specific example, additional preset constraints can be set for virtual resource borrowing scenarios. For instance, in scenarios involving borrowing computing resources, the characteristics of the tasks the borrowing institution needs to perform can be considered, further matching them with the characteristics of the computing resources provided by the lending institution. This improves the accuracy of computing resource borrowing and the efficiency of task execution. The computing equipment provided by the lending institution may have different characteristics, such as suitability for data computation, image processing, or model training. These characteristics can be matched with the task characteristics the borrowing institution needs to perform. For example, a borrowing institution needing to perform an "image processing" task can be matched with a lending institution that provides computing equipment "suitable for image processing," thus determining the appropriate resource lending scheme. Therefore, the preset constraints can also include: for each borrowing institution, the matching degree between the resources borrowed and the task requirements of the borrowing institution is higher than a preset matching degree threshold.

[0057] The embodiments of this application do not limit the specific process for determining the initial resource lending scheme. Optionally, a solver can be used in conjunction with preset constraints to determine the initial resource lending scheme.

[0058] In operation S230, the initial resource lending scheme can be determined as the current resource lending scheme. Then, the current resource lending scheme can be iteratively optimized to gradually reduce the number of resource lending relationships. This can involve iteratively executing preset steps to determine the target lending institution and a corresponding resource lending relationship (including any target borrowing institution and the target amount of resource lent). Further optimization is performed based on preset optimization targets to reduce the target resource amount. Combined with preset constraints, the resource amounts in other resource lending relationships are adjusted so that the updated new resource lending scheme also meets the preset constraints. Accordingly, if the number of target borrowing institutions in any lending institution's new resource lending scheme is less than the number of target borrowing institutions in the current resource lending scheme, it can be determined that the corresponding resource lending relationship has been deleted, thereby reducing at least one resource lending relationship and the number of resource lending relationships. If the target resource amount in the new resource lending scheme is reduced to 0, it can be determined that the corresponding resource lending relationship can be deleted, thereby reducing at least one resource lending relationship and the number of resource lending relationships.

[0059] It should be noted that determining the target resource amount is primarily for the purpose of reducing the target resource amount and minimizing the resource lending relationships associated with it. When adjusting the lending resource amount in the current resource lending plan, it is clear that the target resource amount needs to be reduced.

[0060] Therefore, by operating S230, the number of resource lending relationships can be gradually reduced. The main goal of each iteration can be to reduce at least one resource lending relationship, thereby reducing the difficulty and computational load of reducing the number of resource lending relationships and improving the efficiency of determining resource lending schemes.

[0061] In the preset steps of operation S230, the embodiments of this application do not limit other branch situations. Specifically, it may be a branch situation where the new resource lending plan does not meet the preset update conditions, or a new resource lending plan has not been determined for a long time. The embodiments of this application do not limit the corresponding operations to be performed.

[0062] Optionally, the preset steps may also include at least one of the following: (1) if the new resource lending plan of each lending institution does not meet the preset update conditions, the current resource lending plan of each lending institution before the update is redefined as the current resource lending plan of each lending institution; (2) if no new resource lending plan of each lending institution is determined within the preset time threshold, the current resource lending plan of each lending institution before the update is redefined as the current resource lending plan of each lending institution. In this embodiment, if the new resource lending plan does not meet the preset update conditions, or if no new resource lending plan is determined for a long time, a rollback can be performed, and the current resource lending plan of each lending institution before the update can be redefined as the current resource lending plan of each lending institution, and iterative optimization can be performed again, thereby improving the efficiency and accuracy of determining the resource lending plan.

[0063] In the case of iterative optimization after rollback, the determined target lending institution and / or target resource quantity can be different from the previous iteration cycle, thus allowing for iterative optimization from multiple perspectives and improving the efficiency and accuracy of determining the resource lending scheme. Therefore, optionally, determining the target lending institution can specifically involve identifying any lending institution that was not identified as a target lending institution in the historical cycle or the previous cycle; determining the target resource quantity can involve identifying any resource quantity that was not identified as a target resource quantity in the historical cycle or the previous cycle. For the target lending institution, since it can have multiple resource lending relationships, the same lending institution can be repeatedly identified as a target lending institution in different iteration cycles. As for the target resource quantity, if it is updated to 0 in the current cycle, the corresponding resource lending relationship will not exist in the next cycle, and it will not be repeatedly identified as a target resource quantity in the next cycle; if a rollback occurs in the current cycle, the resource quantity in other resource lending relationships can be identified as the target resource quantity in the next cycle.

[0064] The embodiments of this application do not limit the method of determining the target lending institution. Optionally, any lending institution can be determined as the target lending institution, or a lending institution can be randomly selected as the target lending institution; alternatively, any lending institution with a large number of target borrowing institutions in the current resource lending scheme can be determined as the target lending institution, which is convenient to reduce the number of target borrowing institutions.

[0065] Therefore, optionally, the target lending institution is determined by any of the following: (1) any lending institution in the current resource lending scheme whose number of target borrowing institutions is greater than a preset threshold is determined as the target lending institution; (2) the lending institution in the current resource lending scheme whose number of target borrowing institutions is greater than a preset threshold and whose number of target borrowing institutions is the largest is determined as the target lending institution; (3) among the lending institutions in the current resource lending scheme whose number of target borrowing institutions is greater than a preset threshold, the target lending institution is determined based on the number of target borrowing institutions; wherein, the probability of a lending institution being determined as the target lending institution is positively correlated with the number of target borrowing institutions. This embodiment can determine the target lending institution based on the number of target borrowing institutions in the resource lending scheme, and can optimize by selecting lending institutions with a larger number of target borrowing institutions, which can reduce the difficulty and computational load of reducing the number of resource lending relationships, and can improve the efficiency of reducing the number of resource lending relationships and determining the resource lending scheme. Of course, the target lending institution can also be determined by other methods.

[0066] The embodiments of this application do not limit the specific method of determining the target resource quantity. Optionally, the resource lending relationship between the target lending institution and any target borrowing institution may be randomly selected, and the amount of resources lent therein may be determined as the target resource quantity; alternatively, the selection or determination may be based on the size of the resources lent by the target lending institution; or it may be determined based on the priority of the resource lending relationship. It is understood that the determined target resource quantity is used to reduce the resource quantity during the iterative optimization process, so as to reduce the number of resource lending relationships.

[0067] Optionally, determining the target resource amount lent by the target lending institution to any target borrowing institution may specifically include any of the following: (1) determining the minimum resource amount lent by the target lending institution to the target borrowing institution as the target resource amount; (2) determining the target resource amount based on the resource amount lent by the target lending institution to the target borrowing institution; wherein the probability of the resource amount being determined as the target resource amount is negatively correlated with the size of the resource amount. This embodiment can determine the target resource amount based on the size of the resource amount, and can optimize by selecting the target resource amount with a smaller resource amount as much as possible, which can reduce the difficulty and computational load of reducing the number of resource lending relationships, and improve the efficiency of reducing the number of resource lending relationships and determining the resource lending scheme. Of course, the target resource amount can also be determined in other ways.

[0068] Given a defined target lending institution and target resource quantity, the current resource lending plan can be further optimized. This can be achieved by reducing the target resource quantity and, in conjunction with preset constraints, adjusting the resource quantities in other lending relationships. Specifically, this involves adjusting the resource quantities in the current lending plans of each lending institution to ensure that the target resource quantity is reduced while meeting preset constraints, thereby optimizing and updating the current resource lending plan to obtain a new one.

[0069] The embodiments of this application are not limited to a specific method of obtaining a new resource lending scheme through optimization. Optionally, it may be possible to use a solver to adjust the amount of resources in the current resource lending scheme of each lending institution based on a preset optimization objective and preset constraints, and reduce the target amount of resources.

[0070] The embodiments of this application are not limited to a preset optimization goal. Optionally, the preset optimization goal may include reducing the target resource amount, or minimizing the update magnitude of each optimization update scheme, etc.

[0071] Optionally, the preset optimization objective may include at least one of the following: (1) maximizing the reduction in the target resource amount; (2) minimizing the absolute value of the resource change in other resource amounts besides the target resource amount in the current resource lending schemes of each lending institution; (3) minimizing the number of resource changes in other resource amounts besides the target resource amount in the current resource lending schemes of each lending institution; (4) maximizing the comprehensive change value of the resource amount; the comprehensive change value of the resource amount is positively correlated with the reduction in the target resource amount, negatively correlated with the absolute value of the resource change in other resource amounts besides the target resource amount in the current resource lending schemes of each lending institution, and negatively correlated with the number of resource changes in other resource amounts besides the target resource amount in the current resource lending schemes of each lending institution. This embodiment can improve the efficiency and accuracy of optimizing and updating the resource lending scheme by setting specific preset optimization objectives.

[0072] Among them, maximizing the reduction of the target resource amount means that the target resource amount needs to be reduced as much as possible when optimizing the scheme. Correspondingly, this optimization objective can also be minimizing the target resource amount.

[0073] Minimizing the absolute value of resource changes other than the target resource quantity, and minimizing the number of resource changes among other resources other than the target resource quantity, can be done by minimizing the update magnitude of resource quantities as much as possible when optimizing the scheme.

[0074] It is understandable that the overall change in resource quantity can be correlated with various influencing factors, and thus can be determined using a weighted sum formula. The overall change in resource quantity can be the weighted sum of three items: the reduction in the target resource quantity, the absolute value of the change in other resource quantities besides the target resource quantity in the current lending schemes of various lending institutions, and the change in the quantity of other resource quantities besides the target resource quantity in the current lending schemes of various lending institutions. The reduction in the target resource quantity can have a positive weight; the absolute value of the change in other resource quantities besides the target resource quantity in the current lending schemes of various lending institutions, and the change in the quantity of other resource quantities besides the target resource quantity in the current lending schemes of various lending institutions, can have negative weights.

[0075] Of course, other methods can also be used to determine the comprehensive change value of resource quantity. The embodiments of this application do not limit the specific method of determining the comprehensive change value of resource quantity.

[0076] After updating and optimizing the current resource lending plans of lending institutions based on preset optimization objectives and constraints, new resource lending plans for each institution can be determined. These new plans are determined based on the current plans; however, they cannot be directly adopted as the new current plans at this stage. Subsequently, in operation S230, it can be determined whether the current resource lending plans of each lending institution need to be updated based on preset update conditions.

[0077] The embodiments of this application do not limit the specific content of the preset update conditions. In particular, as the resource lending scheme is updated and optimized, the target resource amount may be reduced, which may cause the resource amount borrowed by some target borrowing institutions to decrease to 0, thereby reducing the number of target borrowing institutions and resource lending relationships.

[0078] Optionally, the preset update condition may include: the number of target borrowing institutions in any new resource lending scheme of any lending institution is less than the number of target borrowing institutions in the current resource lending scheme. Specifically, this could mean that there is any one or at least one lending institution where the number of target borrowing institutions in the new resource lending scheme is less than the number of target borrowing institutions in the current resource lending scheme, thereby reducing the number of resource lending relationships. Therefore, the preset update condition can be determined to be met when the number of resource lending relationships is determined to have decreased. Optionally, the preset update condition may include: the total number of target borrowing institutions in the new resource lending schemes of all lending institutions is less than the total number of target borrowing institutions in the current resource lending schemes of all lending institutions. It is understood that different preset update conditions can be set to determine whether the number of resource lending relationships has decreased.

[0079] Understandably, the preset update conditions can include: a target resource quantity of 0, meaning that after the plan is updated and optimized, the target resource quantity is reduced to 0, and the preset constraints are met, thereby reducing at least one resource lending relationship. Here, a target resource quantity of 0 indicates that in the new resource lending plan of the target lending institution, there is no corresponding resource lending relationship, and no corresponding target borrowing institution. Alternatively, in response to a target resource quantity of 0, the corresponding target borrowing institution or the corresponding resource lending relationship can be deleted from the new resource lending plan of the target lending institution.

[0080] It should be noted that if at least one resource lending relationship or at least one target borrowing institution is reduced in the new resource lending plan, it can be determined that the number of resource lending relationships and target borrowing institutions has decreased, and it can be determined that the current resource lending plan can be updated, and the new resource lending plan of each lending institution can be determined as the new current resource lending plan of each lending institution.

[0081] Of course, the above preset update conditions are for illustrative purposes only, and other preset update conditions can also be set to update the current resource lending scheme.

[0082] The above embodiments explain the specific process of iterative iteration. Through multiple iterations, the number of resource lending relationships and target borrowing institutions can be continuously reduced, thereby achieving continuous optimization of the resource lending scheme.

[0083] Accordingly, the embodiments of this application do not limit the specific conditions for stopping the loop. Optionally, the loop may stop when the loop reaches a preset number of times, or when the number of resource lending relationships is less than a preset number. That is, the loop stops when the resource lending scheme is optimized to meet the requirements, or when further optimization is not possible.

[0084] Optionally, the preset loop stopping condition includes at least one of the following: (1) the number of loops is greater than a preset number threshold; (2) the number of target borrowing institutions in the current resource lending scheme of each lending institution is less than or equal to a preset institution number threshold; (3) the number of lending institutions borrowing resources by each borrowing institution satisfies a preset sparsity constraint. This embodiment can limit specific loop stopping conditions to improve the efficiency of iterative optimization of resource lending schemes.

[0085] Specifically, the preset number threshold for the number of institutions can be the preset number threshold in the above embodiment of determining the target lending institutions, so that the loop can be stopped when the number of target borrowing institutions of each lending institution is not greater than the preset number threshold.

[0086] Furthermore, the embodiments of this application are not limited to the preset sparsity constraint. Specifically, it can be that the number of lending institutions borrowing resources from each borrowing institution is less than a preset threshold for the number of lending institutions. This allows borrowing institutions to reduce the overall complexity of resource lending relationships by borrowing resources from as few lending institutions as possible.

[0087] In an optional embodiment, the control node can invoke a computing node to optimize the resource lending scheme in a preset step. The optimization can be based on preset optimization goals and constraints, thereby reducing the computational load on the control node by utilizing the computing node. Specifically, the control node can assign corresponding preset tasks to the computing node for execution. The computing node can provide feedback on the determined new resource lending scheme, and the control node, based on preset update conditions, determines whether a new current resource lending scheme has been determined. The preset tasks may specifically include: updating the current resource lending scheme of each lending institution based on preset optimization goals, reducing the target resource amount, and obtaining a new resource lending scheme for each lending institution that satisfies preset constraints. Optionally, the control node can also invoke a computing node to execute the entire loop operation.

[0088] Furthermore, considering that new resource lending schemes can be determined by preset optimization objectives based on "target lending institutions and target resource quantities," new resource lending schemes under different branches can be determined separately based on the correspondence between multiple sets of different "target lending institutions and target resource quantities" through preset optimization objectives. Then, the optimal new resource lending scheme can be selected from the new resource lending schemes under different branches.

[0089] Among these, the target lending institutions and target resource quantities can be the same or different, and the target resource quantities can be different.

[0090] For multiple sets of different "target lending institutions and target resource quantities", the distributed characteristics of control nodes and computing nodes can be combined to call multiple different computing nodes in parallel to determine new resource lending schemes for multiple sets of different "target lending institutions and target resource quantities", thereby improving the accuracy, comprehensiveness and efficiency of updating and optimizing resource lending schemes.

[0091] It is understandable that even if the new resource lending scheme determined by some computing nodes does not meet the preset update conditions, or if some computing nodes cannot determine a new resource lending scheme, a rollback can be performed, or a new resource lending scheme determined by other computing nodes can be selected as the new current resource lending scheme.

[0092] Furthermore, the control node can issue corresponding preset tasks to different computing nodes based on different "target lending institutions and target resource amounts" to determine new resource lending schemes. The control node can obtain the different new resource lending schemes determined by different computing nodes, compare them, and select the optimal new resource lending scheme.

[0093] Therefore, optionally, the control node is used to control multiple computing nodes; determine the target lending institution and the target resource amount to be lent by the target lending institution to any target borrowing institution, specifically including: determining the correspondence between different groups of target lending institutions and target resource amounts respectively; calling computing nodes, based on preset optimization objectives, updating the resource amount lent to each target borrowing institution in the current resource lending scheme of each lending institution, reducing the target resource amount, and obtaining new resource lending schemes for each lending institution, wherein the new resource lending schemes of each lending institution satisfy preset constraints, including: calling different computing nodes in parallel, performing the following operations respectively on the correspondence between different groups of target lending institutions and target resource amounts, to obtain a corresponding set of new resource lending schemes: based on preset optimization objectives, updating the resource amount lent to each target borrowing institution in the current resource lending scheme of each lending institution, reducing the target resource amount, and obtaining new resource lending schemes for each lending institution as a corresponding set of new resource lending schemes, wherein the new resource lending schemes of each lending institution satisfy preset constraints; the preset update conditions also include: preset attributes are superior to any other set of new resource lending schemes. This embodiment can determine a set of new resource lending schemes in parallel by using multiple computing nodes based on the correspondence between different groups of target lending institutions and target resource quantities. Then, the optimal new resource lending scheme can be selected, which can improve the accuracy, comprehensiveness and efficiency of updating and optimizing resource lending schemes.

[0094] It is understandable that determining the correspondence between different groups of target lending institutions and target resource quantities can be achieved by separately defining the "resource quantities lent between different lending institutions and borrowing institutions" as target resource quantities, thereby determining the corresponding lending institutions as target lending institutions and obtaining different groups of correspondence.

[0095] The embodiments of this application do not limit the preset update conditions. Among them, for multiple different sets of new resource lending schemes determined by multiple computing nodes (each set of new resource lending schemes contains new resource lending schemes of various lending institutions), the optimal set of new resource lending schemes can be further selected.

[0096] The embodiments of this application do not limit the specific method of selecting the optimal set of new resource lending schemes. Specifically, it can be determined based on preset attributes, that is, selecting a set of new resource lending schemes whose preset attributes are better than any other set of new resource lending schemes or better than all other sets of new resource lending schemes, and determining the new resource lending schemes of each lending institution in it as the new current resource lending schemes of each lending institution. The embodiments of this application do not limit the preset attributes. Optionally, the preset attributes may specifically include at least one of the following: (1) the number of reduced resource lending relationships; (2) the total number of target borrowing institutions in the new resource lending schemes of each lending institution; (3) the total reduction in the number of target borrowing institutions in the new resource lending schemes of each lending institution compared to the number of target borrowing institutions in the current resource lending schemes of the same lending institution.

[0097] It is understandable that the set of new resource lending schemes that minimizes the total number of target borrowing institutions among the new resource lending schemes of various lending institutions can be determined as the optimal set of new resource lending schemes. Alternatively, the set of new resource lending schemes that maximizes the total reduction in the aforementioned number can be determined as the optimal set of new resource lending schemes. Furthermore, the new resource lending schemes of each lending institution in the optimal set of new resource lending schemes can be determined as the new current resource lending schemes of each lending institution.

[0098] The embodiments of this application also provide another method embodiment.

[0099] Figure 3 A flowchart illustrating another resource allocation method according to an embodiment of this application is shown schematically. Figure 3 As shown, the resource allocation method provided in this application embodiment may include operations S310 to S330. The embodiments of this application do not limit the specific entity executing the resource allocation method; it can be executed by any electronic device or any software application.

[0100] In operation S310: determine the set of lending institutions and the set of borrowing institutions; any lending institution is used to lend resources to one or more borrowing institutions.

[0101] In operation S320: Based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as the target of resource lending, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints.

[0102] In operation S330: the initial resource lending plan of each lending institution is determined as the current resource lending plan of each lending institution, and the following preset steps are executed cyclically until a preset loop stop condition is met: determine the target lending institution and determine the target resource amount lent by the target lending institution to any target borrowing institution; based on the preset optimization target, update the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution, reduce the target resource amount, and obtain a new resource lending plan for each lending institution, and the new resource lending plan of each lending institution meets the preset constraint condition; if the new resource lending plan of each lending institution meets the preset update condition, the new resource lending plan of each lending institution is determined as the new current resource lending plan of each lending institution; the preset update condition includes: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

[0103] For an explanation of this method flow, please refer to the explanations of other embodiments.

[0104] Optionally, the following steps are taken: First, determine the target lending institutions and the target resource amount to be lent by each target lending institution to any target borrowing institution. This includes: determining the correspondence between different groups of target lending institutions and target resource amounts; based on a preset optimization objective, updating the resource amount lent to each target borrowing institution in the current resource lending scheme of each lending institution, reducing the target resource amount, and obtaining new resource lending schemes for each lending institution. The new resource lending schemes of each lending institution satisfy preset constraints. Specifically, this may include: performing the following operations in parallel for different groups of target lending institutions and target resource amounts to obtain a corresponding set of new resource lending schemes: based on the preset optimization objective, updating the resource amount lent to each target borrowing institution in the current resource lending scheme of each lending institution, reducing the target resource amount, and obtaining new resource lending schemes for each lending institution as a corresponding set of new resource lending schemes. The new resource lending schemes of each lending institution satisfy preset constraints. The preset update conditions also include: a preset attribute that is superior to any other set of new resource lending schemes.

[0105] For ease of understanding, this application also provides an application embodiment. In the market, there are typically numerous lending institutions that can lend resources, and many differentiated borrowing institutions that need to borrow resources. Furthermore, the resource demands and borrowing rates among these borrowing institutions can vary significantly. For such inter-market resource lending transactions, the scale of resources involved is enormous, and the number of institutions is vast. Therefore, the primary task for the transaction department is to design well-performing resource lending schemes that meet the needs of both lending and borrowing institutions, as well as comply with market transaction norms. The design of resource lending schemes can also be referred to as transaction volume.

[0106] In determining resource lending schemes, at least two necessary technical constraints must be met due to business requirements. First is the consistency constraint, which requires that the weighted average rates allocated to each lending institution be as close as possible (consistent weighted average rates) to ensure fair allocation. Generally, the difference in weighted average rates among lending institutions must not exceed a given threshold, such as 2 basis points. Second is the sparsity constraint, which, to ensure transaction execution efficiency, should not break down orders too much among borrowing institutions, maintaining the sparsity of resource allocation as much as possible; that is, the number of lending institutions for a borrowing institution's resources should not be too large. Several other basic constraints must also be met, such as the amount of resources lent in each resource lending transaction needing to be greater than a lower threshold. However, currently, determining resource lending schemes is usually done manually by the business department. Due to the large number of institutions and complex technical constraints, this method consumes a significant amount of human resources and is inefficient.

[0107] To overcome the problems encountered in existing transaction partitioning processes, and under the premise of meeting the technical constraints of business requirements, an automated and efficient transaction partitioning system is designed. This embodiment designs a novel transaction partitioning system, including an initial solution solving module, an iterative improvement module, and a data processing module. After obtaining relevant data from the business system, the transaction partitioning system solves a linear programming problem to obtain an initial resource allocation scheme. Then, it iteratively corrects the initial solution by performing multiple linear programming solutions, thereby outputting an efficient resource lending scheme with fewer transaction partitioning operations while meeting technical and basic constraints.

[0108] This embodiment breaks down the transaction component task into an initial solution solving module, an iterative improvement module, and a data processing module. Specifically, in the first two modules, this embodiment will... ij Treating X as a continuous variable reduces the complexity of the problem and processes it in the third step to meet accuracy requirements. ij It is the amount of resources (integer variable) that lending institution i lends to borrowing institution j.

[0109] In the initial solution solving module, the transaction component system first obtains data on lending and borrowing institutions from the transaction database. After some basic database processing, the required data is input into the open-source solver according to the established linear programming model to obtain a basic solution that satisfies the consistency constraints.

[0110] The iterative improvement module takes a basic solution of the linear programming model as input, solves a linear programming problem multiple times, reduces the number of transaction splits in the original feasible resource lending scheme, and finally obtains a basic solution that satisfies the sparsity constraint, and further reduces transaction costs.

[0111] In the solution process of the first two modules, due to the difficulty in specifying the accuracy of the results obtained by the linear programming solver, the accuracy is often higher than the actual requirements. The data processing module designs a system of linear integer equations. By solving the solution of this system of equations, the basic solution obtained by the first two modules can be corrected to obtain a feasible solution that meets all requirements.

[0112] During the initial solution solving module and the iterative improvement module, the dynamically changing basic solution always satisfies all specification constraints and gradually satisfies all technical constraints of business requirements, including consistency constraints and sparsity constraints.

[0113] Since directly solving the transaction component problem is an integer programming problem, it presents significant challenges. This embodiment simplifies the problem into a linear programming problem. An open-source solver is used for initial solution, yielding a basic solution that satisfies consistency constraints. Subsequently, the iterative improvement of the resource lending scheme is modeled as a network flow problem. By solving a weighted maximum flow problem, the resource lending scheme is revised, continuously reducing the number of transaction splits to lower transaction costs. This yields a basic solution that meets all business requirements and technical constraints. Finally, after data processing, a resource lending scheme that satisfies all requirements is obtained.

[0114] The overall process is as described above. The automated transaction component system integrates an initial solution module, an iterative improvement module, and a data processing module. This system can automatically process data from business systems and quickly output transaction components that meet technical constraints. The specific process of each step is shown below.

[0115] Initial solution solving module: Take X ij As continuous variables, the model is solved using an open-source solver with time constraints, resulting in an initial basic resource lending scheme that satisfies basic norms and consistency constraints, but has a relatively high number of transaction splits.

[0116] Iterative Improvement Module: This system treats a basic resource lending scheme for the transaction component problem as a bipartite graph network topology. The lower and upper parts represent lending and borrowing institutions, respectively. Directed edges represent the resource lending relationship between lending and borrowing institutions, and flow represents the amount of resources lent, with the direction from the lending institution to the borrowing institution. This basic resource lending scheme must satisfy the following constraints: for any lending institution i, its total outflow equals its held resources; for any borrowing institution j, its total inflow equals its resource borrowing demand.

[0117] A feasible resource lending scheme (satisfying all business rules) also requires the bipartite graph model to additionally satisfy normative and technical constraints. Of course, directly obtaining a feasible resource lending scheme that satisfies all strict constraints is usually very difficult. Therefore, an iterative improvement strategy can be adopted: starting from a basic resource lending scheme (satisfying only basic flow conservation), and gradually adjusting it to eventually approximate a feasible scheme.

[0118] The scheme adjustment here is modeled as a network cyclic flow problem. Specifically, the module selects the borrowing institution with the highest in-degree as the selected borrowing institution and chooses the edge connected to it as the selected edge. The module finds the directed cycle containing this edge through heuristics or a programming solution method, where the flow of each edge in the cycle depends on its resource quantity at its position in the current resource lending scheme and the maximum resource quantity allowed by the corresponding borrowing institution. This cyclic flow can offset the actual resource quantity of the original selected edge, thereby reducing the in-degree of the borrowing institution, i.e., reducing the number of institutional transaction splits.

[0119] Subsequently, the module overlays the graph of the loop flow and the current resource lending scheme to obtain a new resource lending scheme, which reduces at least one edge compared to the current resource lending scheme, i.e. reduces one transaction split, and is a relatively better resource lending scheme.

[0120] Specifically, the algorithm is concerned with resource lending scheme X. ij Given the column with the most non-zero elements, for borrowing institutions that do not meet the splitting requirements or have not achieved the optimal splitting, one edge can be selected as the edge to be eliminated. In the simplest case, by reducing the resource amount of the selected edge and adjusting the resource amounts of other edges accordingly, constraints such as resource balance can be satisfied, thereby reducing at least one edge in the new resource lending scheme.

[0121] In practical implementation, this module models the problem of finding cyclic flows, resulting in a linear programming model for solving the weighted maximum cyclic flow, with the objective function being: ,in For X ij The magnitude of the reverse flow (flow from the borrowing institution to the lending institution), λ ij yes The weight, It is aimed at X ij The magnitude of the positive flow (flow from lending institution to borrowing institution), μ ij yes The module adjusts the hyperparameter λ to determine the weights. ij and μ ij The sign and magnitude of the value control the preference for the edge (increase or decrease the amount of resources), where n is the total number of lending institutions and m is the total number of borrowing institutions.

[0122] Starting with a basic resource lending scheme, we prioritize lending institutions with higher splitting times. Then, from the edges connecting these institutions to all other lending institutions, we select edges with lower resource amounts as target edges to attempt to eliminate. All other edges are unselected.

[0123] For the unselected edge, the hyperparameter λ ij and μ ij All values ​​are set to -1 to avoid traffic passing through and to minimize excessive corrections made to the current resource lending scheme by the cyclic flow obtained from the optimal solution, which could lead to excessive deviation. Only for selected edges will λ be controlled. ij It is positive and λ ij +μ ij A negative value reduces the flow of the selected edge and prevents loops of length 2; or it controls μ. ij It is positive and λ ij +μ ij A negative value increases the flow of the selected edge without creating a loop of length 2. Consistency constraints and upper / lower flow limits are added to ensure the final flow is a cyclic flow rather than a general flow.

[0124] Specifically, the algorithm focuses on the borrowing institution with the highest number of transaction splits, i.e., the borrowing institution with the highest degree, and attempts to reduce the number of splits by the borrowing institution to decrease resource lending costs. The algorithm also focuses on the selected edge; therefore, it assigns a λ value to the selected edge. ij M and μ ij Let M be (-M-1), where M can be a sufficiently large positive number to encourage positive flow to eliminate the selected edge, while avoiding loops of length 2.

[0125] Furthermore, the algorithm incorporates a learning mechanism to adjust edge selection preferences during each iteration. In the first iteration, edge selection is based on the current resource lending scheme; the borrowing institution with the highest in-degree probabilistically selects the edge with the smallest flow or resource amount. After each optimization step, the algorithm modifies the edge weights based on whether the result of that step is acceptable (i.e., whether the optimized resource lending scheme is more effective). If unacceptable, the weight of the currently selected edge is increased to reduce the probability of selecting the same edge in the next step. If accepted, the currently selected edge can be deleted.

[0126] The learning mechanism enables the algorithm to be computed in parallel to a certain extent. At each step, the algorithm can perform different edge selection iterations (random edge selection based on probability) on the same resource lending scheme, and learn multiple modifications to the edge selection strategy from this process. These modifications are then integrated to accelerate the learning of the edge selection strategy and improve optimization efficiency.

[0127] The module solves a linear programming model to obtain new resource lending schemes, evaluates them, and retains the optimal scheme for further iteration. Finally, after a certain number of steps or if the iterations are unsatisfactory, the module outputs a better resource lending scheme with lower transaction costs.

[0128] Data Processing Module: This module processes resource lending schemes that are not precise to the specified resource quantity level. For general resource lending schemes, the schemes are continuous data, and the precision requirement is not high. First, this module truncates the resource lending scheme output by the previous module, retaining only the resource quantity specific to the specified quantity. Then, the remaining resource quantity is redistributed based on the original resource lending scheme (note that the resource lending relationship does not change at this time; this process only fine-tunes the resource quantity in the resource lending relationship), so that it meets the consistency constraint and resource quantity balance constraint (the sum of the resources borrowed by the borrowing institution from each lending institution equals the resource demand of the borrowing institution, and the sum of the resources lent by the lending institution to each borrowing institution equals the resource quantity possessed by the lending institution).

[0129] Based on the above constraints, this module designed a system of integer equations and solved it using an open-source solver. It is easy to obtain a resource lending scheme that meets the accuracy requirements. This allocation scheme also meets all other constraints and is an excellent resource lending scheme.

[0130] This embodiment, by designing an automated and efficient transaction component system, can achieve at least the following technical effects compared to existing manual transaction component methods: (1) Integrating business system data processing and automatic solution, systematically improving transaction component efficiency; (2) The solution obtained by the solution module can guarantee that all technical constraints of business requirements are met; (3) Reducing model complexity, avoiding nonlinear constraints and objectives, improving problem-solving speed, and meeting overall solution efficiency requirements; (4) Efficiently handling accuracy issues, further reducing the complexity of the preceding model; (5) Combining linear programming solution methods to solve transaction component problems more quickly; (6) A high success rate in solving the problem, minimizing the number of times a single transaction is split.

[0131] Corresponding to the above method embodiments, embodiments of this application also provide a resource allocation device. The following will be combined with... Figure 4 The device is described in detail. Figure 4 A schematic block diagram of a resource allocation apparatus according to an embodiment of this application is shown. Figure 4As shown, this embodiment provides a resource allocation device 400 including a first initialization module 410 and a first loop module 420. This device can be applied to a control node, which can be used to control computing nodes.

[0132] The first initialization module 410 is used to determine a set of lending institutions and a set of borrowing institutions; any lending institution is used to lend resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as the target of resource lending, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints. In one embodiment, the first initialization module 410 can be used to execute the operations S210 and S220 described above and related operations, which will not be repeated here.

[0133] The first loop module 420 is used to determine the initial resource lending plan of each lending institution as the current resource lending plan of each lending institution, and to repeatedly execute the following preset steps until a preset loop stop condition is met: determine the target lending institution and determine the target resource amount lent by the target lending institution to any target borrowing institution; call the computing node, based on a preset optimization target, update the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution, reduce the target resource amount, and obtain a new resource lending plan for each lending institution, wherein the new resource lending plan of each lending institution satisfies a preset constraint condition; when the new resource lending plan of each lending institution satisfies a preset update condition, the new resource lending plan of each lending institution is determined as the new current resource lending plan of each lending institution; the preset update condition includes: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan. In one embodiment, the first loop module 420 can be used to execute the operation S230 described above and related operations, which will not be repeated here.

[0134] Optionally, the preset steps also include at least one of the following: if the new resource lending plan of each lending institution does not meet the preset update conditions, the current resource lending plan of each lending institution before the update is redefined as the current resource lending plan of each lending institution; if no new resource lending plan of each lending institution is determined within the preset time threshold, the current resource lending plan of each lending institution before the update is redefined as the current resource lending plan of each lending institution.

[0135] Optionally, the preset optimization objectives include at least one of the following: maximizing the reduction in the target resource quantity; minimizing the absolute value of resource changes in other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution; minimizing the number of resource changes among other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution; maximizing the comprehensive change value of the resource quantity; the comprehensive change value of the resource quantity is positively correlated with the reduction in the target resource quantity, negatively correlated with the absolute value of resource changes in other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution, and negatively correlated with the number of resource changes among other resource quantities besides the target resource quantity in the current resource lending schemes of each lending institution.

[0136] Optionally, the control node is used to control multiple computing nodes; the first loop module 420 is used to: determine the correspondence between different groups of target lending institutions and target resource quantities respectively; call different computing nodes in parallel to perform the following operations on the correspondence between different groups of target lending institutions and target resource quantities respectively, to obtain the corresponding new resource lending scheme set: based on the preset optimization target, update the resource quantity lent to each target borrowing institution in the current resource lending scheme of each lending institution, reduce the target resource quantity, and obtain the new resource lending scheme of each lending institution as the corresponding new resource lending scheme set, and the new resource lending scheme of each lending institution satisfies the preset constraint conditions; the preset update conditions also include: the preset attribute is better than any other new resource lending scheme set.

[0137] Optionally, the first loop module 420 is used to perform any of the following: determine any lending institution in the current resource lending scheme whose number of target borrowing institutions is greater than a preset threshold as a target lending institution; determine the lending institution in the current resource lending scheme whose number of target borrowing institutions is greater than the preset threshold and whose number of target borrowing institutions is the largest as a target lending institution; among the lending institutions in the current resource lending scheme whose number of target borrowing institutions is greater than the preset threshold, determine the target lending institution based on the number of target borrowing institutions; wherein the probability of a lending institution being determined as a target lending institution is positively correlated with the number of target borrowing institutions.

[0138] Optionally, the first loop module 420 is used to perform any of the following: determining the minimum amount of resources lent by the target lending institution to the target borrowing institution as the target resource amount; determining the target resource amount based on the amount of resources lent by the target lending institution to the target borrowing institution; wherein the probability of the resource amount being determined as the target resource amount is negatively correlated with the size of the resource amount.

[0139] Optionally, the preset loop stopping condition includes at least one of the following: the number of loops is greater than a preset number threshold; the number of target borrowing institutions in the current resource lending scheme of each lending institution is less than or equal to a preset institution number threshold; and the number of lending institutions borrowing resources by each borrowing institution meets a preset sparsity constraint.

[0140] Optionally, the preset constraints include at least one of the following: the resource amount of each lending institution is equal to the total resource amount lent to each target borrowing institution in the corresponding resource lending plan; the resource demand of each borrowing institution is equal to the total resource amount lent by each lending institution; the resource amount lent by each lending institution to any borrowing institution is greater than the preset lower limit threshold of the resource amount; for each lending institution, in the corresponding resource lending plan, the proportion of the total resource interest of each target borrowing institution in the resource amount of the target lending institution is within the preset proportion range.

[0141] According to embodiments of this application, any plurality of modules in the first initialization module 410 and the first loop module 420 can be merged into one module, or any one of these modules can be split into multiple modules. Alternatively, at least a portion of the functionality of one or more of these modules can be combined with at least a portion of the functionality of other modules and implemented in one module. According to embodiments of this application, at least one of the first initialization module 410 and the first loop module 420 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented by any other reasonable means of integrating or packaging the circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these. Alternatively, at least one of the first initialization module 410 and the first loop module 420 can be at least partially implemented as a computer program module, which, when run, can perform corresponding functions.

[0142] Corresponding to the above method embodiments, embodiments of this application also provide another resource allocation apparatus. The following will be combined with... Figure 5 The device is described in detail. Figure 5 A schematic block diagram of another resource allocation apparatus according to an embodiment of this application is shown. Figure 5 As shown, the resource allocation device 500 provided in this embodiment includes: a second initial module 510 and a second loop module 520.

[0143] The second initialization module 510 is used to determine a set of lending institutions and a set of borrowing institutions; any lending institution is used to lend resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as the target of resource lending, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints. In one embodiment, the second initialization module 510 can be used to execute the operations S310 and S320 described above and related operations, which will not be repeated here.

[0144] The second loop module 520 is used to determine the initial resource lending plan of each lending institution as the current resource lending plan of each lending institution, and to repeatedly execute the following preset steps until a preset loop stop condition is met: determine the target lending institution and determine the target resource amount lent by the target lending institution to any target borrowing institution; based on a preset optimization target, update the resource amount lent to each target borrowing institution in the current resource lending plan of each lending institution, reduce the target resource amount, and obtain a new resource lending plan for each lending institution, wherein the new resource lending plan of each lending institution satisfies a preset constraint condition; when the new resource lending plan of each lending institution satisfies a preset update condition, the new resource lending plan of each lending institution is determined as the new current resource lending plan of each lending institution; the preset update condition includes: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan. In one embodiment, the second loop module 520 can be used to execute the operation S330 and related operations described above, which will not be repeated here.

[0145] The explanations of the above two device embodiments can be found in the explanations of other embodiments. Any operation in the above method embodiments can be executed by a corresponding module in the device embodiments.

[0146] Figure 6 A block diagram schematically illustrates an electronic device suitable for implementing a resource allocation method according to an embodiment of this application.

[0147] like Figure 6As shown, an electronic device 900 according to an embodiment of this application includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage portion 908 into a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0148] RAM 903 stores various programs and data required for the operation of electronic device 900. Processor 901, ROM 902, and RAM 903 are interconnected via bus 904. Processor 901 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 902 and / or RAM 903. It should be noted that the programs may also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0149] According to embodiments of this application, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to a bus 904. The electronic device 900 may also include one or more of the following components connected to the input / output (I / O) interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 910 as needed so that computer programs read from it can be installed into the storage section 908 as needed.

[0150] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0151] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include ROM 902 and / or RAM 903 and / or one or more memories other than ROM 902 and RAM 903 described above.

[0152] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement a resource allocation method provided in the embodiments of this application.

[0153] When the computer program is executed by the processor 901, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0154] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 909, and / or installed from a removable medium 911. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0155] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 909, and / or installed from the removable medium 911. When the computer program is executed by the processor 901, it performs the functions defined in the system of this application embodiment. According to the embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0156] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0157] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0158] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A resource allocation method, characterized in that, Applied to control nodes; The control node is used to control the computing node; the method includes: Determine the set of lending institutions and the set of borrowing institutions; any lending institution may lend resources to one or more borrowing institutions. Based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints. The initial resource lending plan of each lending institution is determined as the current resource lending plan of each lending institution. The following preset steps are executed repeatedly until the preset loop stop condition is met: Identify the target lending institutions and determine the target amount of resources that the target lending institutions will lend to any target borrowing institution; The computing node is invoked to update the amount of resources lent to each target borrowing institution in the current resource lending plan of each lending institution based on the preset optimization target, thereby reducing the target resource amount and obtaining a new resource lending plan for each lending institution. The new resource lending plan of each lending institution satisfies the preset constraint condition. If the new resource lending plans of each lending institution meet the preset update conditions, the new resource lending plans of each lending institution will be determined as the new current resource lending plans of each lending institution. The preset update conditions include: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

2. The method according to claim 1, characterized in that, The preset steps also include at least one of the following: If the new resource lending plans of each lending institution do not meet the preset update conditions, the current resource lending plans of each lending institution before the update will be redefined as the current resource lending plans of each lending institution. If no new resource lending plan is determined for each lending institution within the preset time threshold, the current resource lending plan of each lending institution before the update will be re-determined as the current resource lending plan of each lending institution.

3. The method according to claim 1, characterized in that, The preset optimization objective includes at least one of the following: Maximize the reduction in the amount of resources required to reach the target resource quantity; In the current resource lending plans of each lending institution, minimize the absolute value of the resource change in other resource quantities besides the target resource quantity; In the current resource lending plans of each lending institution, minimize the number of resource quantities that change among other resource quantities besides the target resource quantity; Maximize the overall change value of resource quantity; the overall change value of resource quantity is positively correlated with the amount of resource reduction of the target resource quantity, negatively correlated with the absolute value of the resource change of other resource quantities besides the target resource quantity in the current resource lending plan of each lending institution, and negatively correlated with the amount of resource quantity change among other resource quantities besides the target resource quantity in the current resource lending plan of each lending institution.

4. The method according to claim 1, characterized in that, The control node is used to control multiple computing nodes; The step of determining the target lending institution and determining the target resource amount lent by the target lending institution to any target borrowing institution includes: determining the correspondence between different groups of target lending institutions and target resource amounts respectively; The process involves calling the computing node to update the resource lending amount for each target borrowing institution in the current resource lending plan of each lending institution based on a preset optimization target, thereby reducing the target resource amount and obtaining a new resource lending plan for each lending institution. The new resource lending plan for each lending institution satisfies the preset constraints, including: Different computing nodes are invoked in parallel to perform the following operations on the correspondence between different groups of target lending institutions and target resource amounts, respectively, to obtain a set of corresponding new resource lending schemes: Based on the preset optimization target, the resource amount lent to each target borrowing institution in the current resource lending scheme of each lending institution is updated, the target resource amount is reduced, and the new resource lending scheme of each lending institution is obtained as the corresponding set of new resource lending schemes, wherein the new resource lending scheme of each lending institution satisfies the preset constraint conditions; The preset update conditions also include: preset attributes are superior to any other new resource lending scheme set.

5. The method according to claim 1, characterized in that, The determination of the target lending institution includes any one of the following: Any lending institution in the current resource lending scheme whose number of target borrowing institutions exceeds a preset threshold is identified as a target lending institution; The lending institution with the largest number of target borrowing institutions in the current resource lending plan that exceeds the preset threshold is identified as the target lending institution. In the current resource lending scheme, among the lending institutions whose number of target borrowing institutions exceeds a preset threshold, the target lending institutions are determined based on the number of target borrowing institutions; the probability of a lending institution being determined as a target lending institution is positively correlated with the number of target borrowing institutions.

6. The method according to claim 1, characterized in that, Determining the target resource amount lent by the target lending institution to any target borrowing institution includes any one of the following: The minimum amount of resources that the target lending institution lends to the target borrowing institution is defined as the target resource amount; Among the resources lent by the target lending institution to the target borrowing institution, the target resource quantity is determined based on the resource quantity; wherein, the probability of a resource quantity being determined as the target resource quantity is negatively correlated with the size of the resource quantity.

7. The method according to claim 1, characterized in that, The preset loop stop condition includes at least one of the following: The number of iterations exceeds a preset threshold. The number of target borrowing institutions in the current resource lending plan of each lending institution is less than or equal to the preset threshold number of institutions; The number of lending institutions for each borrowing institution meets the preset sparsity constraint.

8. The method according to claim 1, characterized in that, The preset constraints include at least one of the following: The amount of resources for each lending institution is equal to the total amount of resources lent to each target borrowing institution in the corresponding resource lending plan; The resource demand of each borrowing institution is equal to the total amount of resources lent by all lending institutions. The amount of resources lent by each lending institution to any borrowing institution exceeds the preset lower limit threshold for resource amount; For each lending institution, in the corresponding resource lending plan, the proportion of the total resource interest of each target borrowing institution in the resource amount of the target lending institution is within the preset proportion range.

9. A resource allocation method, characterized in that, The method includes: Determine the set of lending institutions and the set of borrowing institutions; any lending institution may lend resources to one or more borrowing institutions. Based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints. The initial resource lending plan of each lending institution is determined as the current resource lending plan of each lending institution. The following preset steps are executed repeatedly until the preset loop stop condition is met: Identify the target lending institutions and determine the target amount of resources that the target lending institutions will lend to any target borrowing institution; Based on the preset optimization target, the amount of resources lent to each target borrowing institution in the current resource lending plan of each lending institution is updated, the target resource amount is reduced, and a new resource lending plan for each lending institution is obtained, wherein the new resource lending plan of each lending institution satisfies the preset constraint condition. If the new resource lending plans of each lending institution meet the preset update conditions, the new resource lending plans of each lending institution will be determined as the new current resource lending plans of each lending institution. The preset update conditions include: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

10. A resource allocation device, characterized in that, Applied to control nodes; The control node is used to control the computing node; the device includes: The first initial module is used to determine a set of lending institutions and a set of borrowing institutions; any lending institution is used to lend resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints. The first loop module is used to determine the initial resource lending plan of each lending institution as the current resource lending plan of each lending institution, and to repeatedly execute the following preset steps until the preset loop stop condition is met: Identify the target lending institutions and determine the target amount of resources that the target lending institutions will lend to any target borrowing institution; The computing node is invoked to update the amount of resources lent to each target borrowing institution in the current resource lending plan of each lending institution based on the preset optimization target, thereby reducing the target resource amount and obtaining a new resource lending plan for each lending institution. The new resource lending plan of each lending institution satisfies the preset constraint condition. If the new resource lending plans of each lending institution meet the preset update conditions, the new resource lending plans of each lending institution will be determined as the new current resource lending plans of each lending institution. The preset update conditions include: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

11. A resource allocation device, characterized in that, The device includes: The second initial module is used to determine a set of lending institutions and a set of borrowing institutions; any lending institution is used to lend resources to one or more borrowing institutions; based on the set of lending institutions and the set of borrowing institutions, an initial resource lending plan is determined for each lending institution in the set of lending institutions; the resource lending plan includes one or more target borrowing institutions as resource lending targets, and the amount of resources lent to each target borrowing institution; the determined initial resource lending plan for each lending institution satisfies preset constraints. The second loop module is used to determine the initial resource lending plan of each lending institution as the current resource lending plan of each lending institution, and to repeatedly execute the following preset steps until the preset loop stop condition is met: Identify the target lending institutions and determine the target amount of resources that the target lending institutions will lend to any target borrowing institution; Based on the preset optimization target, the amount of resources lent to each target borrowing institution in the current resource lending plan of each lending institution is updated, the target resource amount is reduced, and a new resource lending plan for each lending institution is obtained, wherein the new resource lending plan of each lending institution satisfies the preset constraint condition. If the new resource lending plans of each lending institution meet the preset update conditions, the new resource lending plans of each lending institution will be determined as the new current resource lending plans of each lending institution. The preset update conditions include: the number of target borrowing institutions in the new resource lending plan of any lending institution is less than the number of target borrowing institutions in the current resource lending plan.

12. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 9.

13. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 9.

14. A computer program product comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 9.

Images