Federated learning task management method and system

Abstract

The present invention discloses a method and system for managing federated learning tasks, which are applied to the field of blockchain technology. The method includes: a management platform receives a federated learning task initiated by a client of a first tenant, and the federated learning task carries task information; the management platform initiates a federated learning task to a blockchain; the management platform receives federated learning task progress information and federated learning task result information fed back by the blockchain, and the federated learning task progress information is sent to the blockchain when the client of the first tenant and the client of the second tenant perform joint modeling, and the second tenant starts to perform joint modeling after receiving the federated learning task initiated by the client of the first tenant. The present invention can perform federated learning task management, and can facilitate user use and effectively constrain the behavior of participants on the basis of achieving data security and reliability.

Description

Technical Field

[0001] The present invention relates to the field of blockchain technology, and in particular to a federated learning task management method and system. Background Art

[0002] In recent years, artificial intelligence has been booming, setting off waves of development, from face recognition to AlphaGo vs. human Go players, to driverless cars, and precision marketing, which has been widely used. AI has gradually entered every aspect of people's lives. Of course, some people are overhyping it, which leads to misunderstandings about AI: AI can do everything. While chasing AI, people overlook the fact that AI is supported by data, and a lot of high-quality data.

[0003] In real life, except for a few giant companies, most companies have problems with small data volume and poor data quality, which are not enough to support the realization of artificial intelligence technology. At the same time, the regulatory environment at home and abroad is gradually strengthening data protection, so the free flow of data under the premise of safety and compliance has become a general trend. From the perspective of users and enterprises, the data owned by commercial companies often has huge potential value. Two companies or even departments between companies have to consider the exchange of interests. Often these institutions will not provide their own data to other companies for single aggregation, resulting in data often appearing in the form of islands even within the same company.

[0004] Based on the above three points of insufficient support for implementation, not allowing rough exchange, and unwillingness to contribute value, a large number of data islands and privacy protection issues have now emerged, and federated learning has come into being. Federated learning is a machine learning framework that can effectively help multiple institutions use data and conduct machine learning modeling while meeting the requirements of user privacy protection, data security, and government regulations. Therefore, federated learning is sometimes directly called federated modeling. Federated learning is essentially a distributed machine learning technology, or a machine learning framework. The goal of federated learning is to achieve joint modeling and improve the effect of AI models on the basis of ensuring data privacy security and legal compliance. At present, the federated learning technology framework is represented by FATE. FATE (Federated AI Technology Enabler) is an open source project initiated by the AI ​​department of WeBank, which provides a reliable secure computing framework for the federated learning ecosystem. The FATE project uses multi-party secure computing (MPC) and homomorphic encryption (HE) technology to build the underlying secure computing protocol to support secure computing of different types of machine learning, including logistic regression, tree-based algorithms, deep learning, and transfer learning.

[0005] Federated learning has attracted widespread attention as an emerging technology. Currently, there are two main existing technologies: centralized scheduling of model training through a centralized system, or fully distributed peer-to-peer connection. However, centralized systems are not easily accepted by data providers and pose data security risks. Although distributed systems are more secure, they are difficult to deploy, complex to configure, and have a high threshold for users to use. Summary of the invention

[0006] The embodiment of the present invention provides a method for managing federated learning tasks, which is used to manage federated learning tasks. It can facilitate user use and effectively constrain the behavior of participants on the basis of achieving data security and reliability. The method includes:

[0007] The management platform receives a federated learning task initiated by a client of the first tenant, wherein the federated learning task carries task information;

[0008] The management platform initiates the federated learning task to the blockchain;

[0009] The management platform receives the federated learning task progress information and federated learning task result information fed back by the blockchain. The federated learning task result information is sent to the blockchain after the first tenant's client and the second tenant's client complete the joint modeling. The federated learning task progress information is sent to the blockchain when the first tenant's client and the second tenant's client perform joint modeling. The second tenant's client starts the joint modeling after receiving the federated learning task initiated by the first tenant's client. After the first tenant's client receives the federated learning task initiated by the management platform on the blockchain, it initiates the federated learning task to the corresponding second tenant's client according to the task information.

[0010] The embodiment of the present invention provides a management platform for managing federated learning tasks, which can facilitate user use and effectively constrain the behavior of participants on the basis of achieving data security and reliability. The management platform includes:

[0011] A task receiving module, used to receive a federated learning task initiated by a client of the first tenant, wherein the federated learning task carries task information;

[0012] A task initiation module, used to initiate the federated learning task to the blockchain;

[0013] An information receiving module is used for the management platform to receive federated learning task progress information and federated learning task result information fed back by the blockchain. The federated learning task result information is sent to the blockchain after the first tenant's client and the second tenant's client complete joint modeling. The federated learning task progress information is sent to the blockchain when the first tenant's client and the second tenant's client perform joint modeling. The second tenant's client starts joint modeling after receiving the federated learning task initiated by the first tenant's client. After the first tenant's client receives the federated learning task initiated by the management platform on the blockchain, it initiates a federated learning task to the corresponding second tenant's client according to the task information.

[0014] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the federated learning task management method.

[0015] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program for executing the above-mentioned federated learning task management method.

[0016] In the embodiment of the present invention, the management platform receives the federated learning task initiated by the client of the first tenant, and the federated learning task carries the task information; the management platform initiates the federated learning task to the blockchain; the management platform receives the federated learning task progress information and the federated learning task result information fed back by the blockchain, and the federated learning task result information is sent to the blockchain by the client of the first tenant and the client of the second tenant after completing the joint modeling, and the federated learning task progress information is sent to the blockchain when the client of the first tenant and the client of the second tenant perform joint modeling, and the client of the second tenant receives the federated learning task initiated by the client of the first tenant and starts the joint modeling, and the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, and initiates the federated learning task to the corresponding client of the second tenant according to the task information. In the embodiment of the present invention, the federated learning scheduled by the blockchain can realize data security and reliability, and can be convenient for users to use, and because the blockchain is used for data storage, the behavior of the participants can be effectively constrained, and after a dispute occurs, the responsibility is determined according to the data on the chain.

[0017] The embodiment of the present invention provides a method for managing federated learning tasks, which is used to manage federated learning tasks. It can facilitate user use and effectively constrain the behavior of participants on the basis of achieving data security and reliability. The method includes:

[0018] The client of the first tenant initiates a federated learning task to the management platform, wherein the federated learning task carries task information;

[0019] After the first tenant's client receives the federated learning task initiated by the management platform on the blockchain, it initiates the federated learning task to the corresponding second tenant's client according to the task information. The management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the first tenant's client;

[0020] After the second tenant's client receives the federated learning task, the first tenant's client and the second tenant's client start joint modeling and send the federated learning task progress information to the blockchain;

[0021] After completing the joint modeling, the client of the first tenant and the client of the second tenant send the federated learning task result information to the blockchain.

[0022] The embodiment of the present invention provides a client for managing federated learning tasks, which can facilitate user use and effectively constrain the behavior of participants on the basis of achieving data security and reliability. The client includes:

[0023] A first task initiating module, used for a client of a first tenant to initiate a federated learning task to a management platform, wherein the federated learning task carries task information;

[0024] The second task initiation module is used for the client of the first tenant to initiate the federated learning task to the client of the corresponding second tenant according to the task information after the blockchain receives the federated learning task initiated by the management platform. The management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the client of the first tenant;

[0025] A joint modeling module, configured to enable the first tenant's client and the second tenant's client to start joint modeling and send the progress information of the federated learning task to the blockchain after the second tenant's client receives the federated learning task;

[0026] The information sending module is used for the client of the first tenant and the client of the second tenant to send the progress information of the federated learning task to the blockchain when performing joint modeling, and to send the result information of the federated learning task to the blockchain after completing the joint modeling.

[0027] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the federated learning task management method.

[0028] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program for executing the above-mentioned federated learning task management method.

[0029] In the embodiment of the present invention, the client of the first tenant initiates a federated learning task to the management platform, and the federated learning task carries task information; after the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, the client of the first tenant initiates the federated learning task to the corresponding client of the second tenant according to the task information, and the management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the client of the first tenant; after the client of the second tenant receives the federated learning task, the client of the first tenant and the client of the second tenant start joint modeling and send the progress information of the federated learning task to the blockchain; after completing the joint modeling, the client of the first tenant and the client of the second tenant send the result information of the federated learning task to the blockchain. In the embodiment of the present invention, the federated learning scheduled by the blockchain can realize data security and reliability, and is convenient for users to use. Moreover, since the blockchain is used for data storage, the behavior of the participants can be effectively constrained. After a dispute occurs, the responsibility is determined according to the data on the chain.

[0030] An embodiment of the present invention provides a federated learning task management system for managing federated learning tasks. It can facilitate user use and effectively constrain the behavior of participants on the basis of achieving data security and reliability. The system includes: the above-mentioned management platform and the above-mentioned client. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0032] Figure 1 Schematic diagram of a method for managing federated learning tasks in an embodiment of the present invention;

[0033] Figure 2 Schematic diagram of another federated learning task management method in an embodiment of the present invention;

[0034] Figure 3 Schematic diagram of another federated learning task management method in an embodiment of the present invention;

[0035] Figure 4 Schematic diagram of another federated learning task management method in an embodiment of the present invention;

[0036] Figure 5 Schematic diagram of another federated learning task management method in an embodiment of the present invention;

[0037] Figure 6Schematic diagram of another federated learning task management method in an embodiment of the present invention;

[0038] Figure 7 Schematic diagram of another federated learning task management method in an embodiment of the present invention;

[0039] Figure 8 Schematic diagram of another federated learning task management method in an embodiment of the present invention;

[0040] Figure 9-10 This is a schematic diagram of a method for managing federated learning tasks in a specific embodiment of the present invention;

[0041] Fig.11 This is a structural diagram of a federated learning task management device in an embodiment of the present invention;

[0042] Fig.12 This is a structural diagram of the management platform in an embodiment of the present invention;

[0043] Fig.13 This is a client structure diagram in an embodiment of the present invention;

[0044] Fig.14 It is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0046] As mentioned above, there are two main methods in the existing technology, centralized management and scheduling system or distributed peer-to-peer system: 1. The centralized system is not easily accepted by data providers and there are data security risks. The schematic diagram is as follows. The computing nodes are directly scheduled by the management center, and the data providers (computing nodes) are at risk of data leakage and cannot be controlled. 2. Although the distributed peer-to-peer system has more secure data, the identity data is managed separately, which is not only inconvenient to use, but also difficult to determine the responsibility once a dispute occurs in the federated modeling learning between two or more parties.

[0047] In order to manage federated learning tasks, the embodiment of the present invention provides a method for managing federated learning tasks, which can facilitate user use and effectively constrain the behavior of participants on the basis of achieving data security and reliability. Figure 1 As shown, the method may include:

[0048] Step 101: The management platform receives a federated learning task initiated by a client of a first tenant, wherein the federated learning task carries task information;

[0049] Step 102: The management platform initiates the federated learning task to the blockchain;

[0050] Step 103: The management platform receives the federated learning task progress information and the federated learning task result information fed back by the blockchain. The federated learning task result information is sent to the blockchain after the client of the first tenant and the client of the second tenant complete the joint modeling. The federated learning task progress information is sent to the blockchain when the client of the first tenant and the client of the second tenant perform joint modeling. The second tenant starts the joint modeling after the client receives the federated learning task initiated by the client of the first tenant. After the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, the client of the first tenant initiates the federated learning task to the corresponding client of the second tenant according to the task information.

[0051] Depend on Figure 1 As shown, it can be known that the embodiment of the present invention receives the federated learning task initiated by the client of the first tenant through the management platform, and the federated learning task carries the task information; the management platform initiates the federated learning task to the blockchain; the management platform receives the federated learning task progress information and the federated learning task result information fed back by the blockchain, and the federated learning task result information is sent to the blockchain by the client of the first tenant and the client of the second tenant after completing the joint modeling, and the federated learning task progress information is sent to the blockchain when the client of the first tenant and the client of the second tenant perform joint modeling, and the client of the second tenant receives the federated learning task initiated by the client of the first tenant and starts the joint modeling. After the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, the client of the first tenant initiates the federated learning task to the corresponding client of the second tenant according to the task information. The embodiment of the present invention can realize data security and reliability through the federated learning scheduled by the blockchain, and it can also be convenient for users to use, and because the blockchain is used for data storage, it can effectively constrain the behavior of the participants, and after a dispute occurs, the responsibility is determined according to the data on the chain.

[0052] In the embodiment, the task information includes: initiator information, participant information, data hash list information, task configuration information, one or any combination thereof.

[0053] In the embodiment, Figure 2 As shown, the federated learning task management method also includes:

[0054] Step 201: The management platform receives a first blockchain account registration request sent by a client of a first tenant;

[0055] Step 202: The management platform sends the information to be filled in to the client of the first tenant according to the first blockchain account registration request;

[0056] Step 203: The management platform receives first registration information fed back by the client of the first tenant according to the information to be filled in;

[0057] Step 204: The management platform initiates a first identity information binding request to the blockchain, where the first identity information binding request carries the first registration information.

[0058] In this embodiment, the first registration information includes: identity information, public key information and computing node information of the first tenant.

[0059] In this embodiment, the blockchain registers the first local data of the client of the first tenant, and the first local data includes: one or any combination of the data hash information, description information, and attribution information of the first tenant.

[0060] In the embodiment, Figure 3 As shown, the federated learning task management method also includes:

[0061] Step 301: The management platform receives a second blockchain account registration request sent by a client of a second tenant;

[0062] Step 302: The management platform sends the information to be filled in to the client of the second tenant according to the second blockchain account registration request;

[0063] Step 303: The management platform receives the second registration information fed back by the client of the second tenant according to the information to be filled in;

[0064] Step 304: The management platform initiates a second identity information binding request to the blockchain, where the second identity information binding request carries the second registration information.

[0065] In this embodiment, the second registration information includes: identity information, public key information and computing node information of the second tenant.

[0066] In this embodiment, the blockchain registers the second local data of the client of the second tenant, and the second local data includes: one or any combination of data hash information, description information, and attribution information of the second tenant.

[0067] In order to manage federated learning tasks, the embodiment of the present invention provides another method for managing federated learning tasks, which is convenient for users to use and effectively constrains the behavior of participants on the basis of achieving data security and reliability. Figure 4 As shown, the method may include:

[0068] Step 401: The client of the first tenant initiates a federated learning task to the management platform, wherein the federated learning task carries task information;

[0069] Step 402: After the first tenant's client receives the federated learning task initiated by the management platform on the blockchain, the first tenant's client initiates the federated learning task to the corresponding second tenant's client according to the task information. The management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the first tenant's client.

[0070] Step 403: After the second tenant's client receives the federated learning task, the first tenant's client and the second tenant's client start joint modeling and send the federated learning task progress information to the blockchain;

[0071] Step 404: After completing the joint modeling, the client of the first tenant and the client of the second tenant send the federated learning task result information to the blockchain.

[0072] Depend on Figure 4 As shown, it can be known that in the embodiment of the present invention, the client of the first tenant initiates a federated learning task to the management platform, and the federated learning task carries task information; after the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, the client of the first tenant initiates the federated learning task to the corresponding client of the second tenant according to the task information, and the management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the client of the first tenant; after the client of the second tenant receives the federated learning task, the client of the first tenant and the client of the second tenant start joint modeling and send the progress information of the federated learning task to the blockchain; after completing the joint modeling, the client of the first tenant and the client of the second tenant send the result information of the federated learning task to the blockchain. The embodiment of the present invention can realize data security and reliability through the federated learning scheduled by the blockchain, and it can also be convenient for users to use, and because the blockchain is used for data storage, the behavior of the participants can be effectively constrained, and after a dispute occurs, the responsibility is determined according to the data on the chain.

[0073] In the embodiment, the task information includes: initiator information, participant information, data hash list information, task configuration information, one or any combination thereof.

[0074] In the embodiment, Figure 5 As shown, the federated learning task management method also includes:

[0075] Step 501: The client of the first tenant sends a first blockchain account registration request to the management platform;

[0076] Step 502: The client of the first tenant receives the information to be filled in sent by the management platform according to the first registration blockchain account request;

[0077] Step 503: The client of the first tenant sends the first registration information corresponding to the information to be filled in to the management platform. The first registration information is used to be carried in the first identity information binding request initiated by the management platform to the blockchain.

[0078] In the embodiment, Figure 6 As shown, Figure 5 The federated learning task management method further includes:

[0079] Step 504: The client of the first tenant registers first local data in the blockchain, where the first local data includes: one or any combination of data hash information, description information, and attribution information of the first tenant.

[0080] In the embodiment, Figure 7 As shown, the federated learning task management method also includes:

[0081] Step 701: The client of the second tenant sends a second blockchain account registration request to the management platform;

[0082] Step 702: The client of the second tenant receives the information to be filled in sent by the management platform according to the second registration blockchain account request;

[0083] Step 703: The client of the second tenant sends the second registration information corresponding to the information to be filled in to the management platform. The second registration information is used to be carried in the second identity information binding request initiated by the management platform to the blockchain.

[0084] In the embodiment, Figure 8 As shown, Figure 7 The federated learning task management method further includes:

[0085] Step 704: The client of the second tenant registers second local data in the blockchain, where the second local data includes: one or any combination of data hash information, description information, and attribution information of the second tenant.

[0086] A specific example is given below to illustrate the specific application of the method for managing federated learning tasks in the embodiment of the present invention. Figure 9-10 As shown, the federated learning task management is performed in the following steps:

[0087] 1. Identity registration: Each tenant accesses the management platform through a browser and fills in the registration information, which includes the identity information, public key information, and computing node information required to apply for a blockchain account.

[0088] 2. The management platform initiates an identity information binding request to the blockchain, including identity information, public key information, and computing node information.

[0089] 3. Data registration: Each tenant can register local data on the blockchain. The main contents include data hash information, description information, ownership information, etc.

[0090] 4. The management platform receives blockchain notifications and obtains the data directory synchronously.

[0091] 5. Task initiation: The tenant of institution 1 logs in to the management platform through the browser and selects the data of institution 2 for joint modeling. This task will also be carried out through the blockchain. The task content should include the initiator information, the participant information, the data hash list information, and the configuration information required for the task. After institution 1 synchronizes the blockchain notification, it is informed that the tenant of this institution is initiating the task. By loading the configuration, it initiates the modeling task to the designated participant. The computing nodes of institution 2 passively accept the task and actively participate. The modeling process generally takes a long time and has many interactions. The computing nodes of institutions 1 and 2 will actively report the task progress to the blockchain until the modeling is completed. The management platform can obtain the task progress and results in real time through the blockchain notification. The tenant of institution 1 (the task initiator) can view it through the management platform. During the whole process, the computing nodes and the management platform interact with each other through the blockchain, and there will be no physical connection, which avoids the risk of data leakage and solves the data security problem.

[0092] The embodiment of the present invention realizes the corresponding service functions of identity management, data management (data market), and task management through blockchain smart contracts; the management platform is responsible for visual configuration and use. It may include basic functions such as task initiation, task status display, and task result display. User information, data market, task status, and task results all come from the blockchain. The computing node only interacts with the blockchain, registers the data on the blockchain, accepts blockchain task scheduling, and feeds back the task status to the blockchain. Using the blockchain as a medium, the identity management, data management, and task management functions are implemented on the management platform. It enhances the trust of the nodes participating in federated learning and facilitates implementation. Data registration and task management are all carried out through the blockchain, the entire information is transparent and reliable, and the computing node can directly respond to related events. The corresponding relationship between identity, data, and tasks is stored in the blockchain, and the entire process can be checked and traced.

[0093] Based on the same inventive concept, the embodiment of the present invention also provides a management platform, as described in the following embodiments. Since the principles of solving these problems are similar to the federated learning task management method, the implementation of the management platform can refer to the implementation of the method, and the repeated parts will not be repeated.

[0094] Fig.11 : is a structural diagram of a federated learning task management device in an embodiment of the present invention. Fig.11 As shown, the management platform 1100 includes:

[0095] The task receiving module 1101 is used to receive a federated learning task initiated by a client of the first tenant, wherein the federated learning task carries task information;

[0096] A task initiating module 1102, used to initiate the federated learning task to the blockchain;

[0097] The information receiving module 1103 is used for the management platform to receive the federated learning task progress information and federated learning task result information fed back by the blockchain. The federated learning task result information is sent to the blockchain after the client of the first tenant and the client of the second tenant complete the joint modeling. The federated learning task progress information is sent to the blockchain when the client of the first tenant and the client of the second tenant perform joint modeling. The second tenant starts the joint modeling after the client receives the federated learning task initiated by the client of the first tenant. After the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, it initiates the federated learning task to the corresponding client of the second tenant according to the task information.

[0098] In summary, the embodiment of the present invention receives the federated learning task initiated by the client of the first tenant through the management platform, and the task information is carried in the federated learning task; the management platform initiates the federated learning task to the blockchain; the management platform receives the federated learning task progress information and the federated learning task result information fed back by the blockchain, and the federated learning task result information is sent to the blockchain by the client of the first tenant and the client of the second tenant after completing the joint modeling, and the federated learning task progress information is sent to the blockchain when the client of the first tenant and the client of the second tenant perform joint modeling, and the client of the second tenant receives the federated learning task initiated by the client of the first tenant and starts the joint modeling. After the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, the client of the first tenant initiates the federated learning task to the corresponding client of the second tenant according to the task information. The embodiment of the present invention can realize data security and reliability through the federated learning scheduled by the blockchain, and it can also be convenient for users to use, and because the blockchain is used for data storage, the behavior of the participants can be effectively constrained, and after a dispute occurs, the responsibility is determined according to the data on the chain.

[0099] Based on the same inventive concept, the embodiment of the present invention also provides a client, as described in the following embodiment. Since the principles of solving these problems are similar to the federated learning task management method, the implementation of the client can refer to the implementation of the method, and the repeated parts will not be repeated.

[0100] Fig.12 is a structural diagram of the client in an embodiment of the present invention, such as Fig.12 As shown, the client 1200 includes:

[0101] A first task initiating module 1201, used for a client of a first tenant to initiate a federated learning task to a management platform, wherein the federated learning task carries task information;

[0102] The second task initiating module 1202 is used for the client of the first tenant to initiate the federated learning task to the client of the corresponding second tenant according to the task information after the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain. The management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the client of the first tenant;

[0103] The joint modeling module 1203 is used for the first tenant's client and the second tenant's client to start joint modeling and send the federated learning task progress information to the blockchain after the second tenant's client receives the federated learning task;

[0104] The information sending module 1204 is used for the client of the first tenant and the client of the second tenant to send the progress information of the federated learning task to the blockchain when performing joint modeling, and to send the result information of the federated learning task to the blockchain after completing the joint modeling.

[0105] In summary, the embodiment of the present invention initiates a federated learning task to the management platform through the client of the first tenant, and the federated learning task carries task information; after the client of the first tenant receives the federated learning task initiated by the management platform on the blockchain, it initiates the federated learning task to the corresponding client of the second tenant according to the task information, and the management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the client of the first tenant; after the client of the second tenant receives the federated learning task, the client of the first tenant and the client of the second tenant start joint modeling and send the progress information of the federated learning task to the blockchain; after completing the joint modeling, the client of the first tenant and the client of the second tenant send the result information of the federated learning task to the blockchain. The embodiment of the present invention can realize data security and reliability through the federated learning scheduled by the blockchain, and it can also be convenient for users to use, and because the blockchain is used for data storage, it can effectively constrain the behavior of the participants, and after a dispute occurs, the responsibility is determined according to the data on the chain.

[0106] Based on the same inventive concept, the embodiment of the present invention also provides a federated learning task management system, as described in the following embodiments. Since the principles of solving these problems are similar to the federated learning task management method, the implementation of the system can refer to the implementation of the method, and the repeated parts will not be repeated.

[0107] Fig.13 : is a structural diagram of a federated learning task management system in an embodiment of the present invention. Fig.13As shown, the federated learning task management system includes: a management platform 1100 and a client 1200 .

[0108] Based on the above invention concept, Fig.14 As shown, the present invention also proposes a computer device 1400, including a memory 1410, a processor 1420, and a computer program 1430 stored in the memory 1410 and executable on the processor 1420, wherein the processor 1420 implements the aforementioned federated learning task management method when executing the computer program 1430.

[0109] Based on the aforementioned inventive concept, the present invention proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the aforementioned federated learning task management method is implemented.

[0110] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0111] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0114] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for managing federated learning tasks, characterized in that: include: The management platform receives a federated learning task initiated by a client of the first tenant, wherein the federated learning task carries task information; The management platform initiates the federated learning task to the blockchain; The management platform receives the federated learning task progress information and federated learning task result information fed back by the blockchain, wherein the federated learning task result information is sent to the blockchain by the client of the first tenant and the client of the second tenant after completing the joint modeling, and the federated learning task progress information is sent to the blockchain by the client of the first tenant and the client of the second tenant when performing joint modeling, and the second tenant starts the joint modeling after the client of the second tenant receives the federated learning task initiated by the client of the first tenant, and the client of the first tenant initiates the federated learning task to the corresponding client of the second tenant according to the task information after the blockchain receives the federated learning task initiated by the management platform; wherein the client of the first tenant and the client of the second tenant actively report the task progress to the blockchain until the modeling is completed; the management platform obtains the task progress and results in real time through the blockchain notification; the tenant of the task initiator views the task progress and results through the management platform; Also includes: The management platform receives a first registration blockchain account request sent by a client of the first tenant; The management platform sends the information to be filled in to the client of the first tenant according to the first blockchain account registration request; The management platform receives the first registration information fed back by the client of the first tenant according to the information to be filled in; wherein each tenant accesses the management platform through a browser and fills in the registration information, including the identity information, public key information, and computing node information required for applying for a blockchain account; The management platform initiates a first identity information binding request to the blockchain, where the first identity information binding request carries the first registration information.

2. The method for managing federated learning tasks according to claim 1, wherein: The task information includes: initiator information, participant information, data hash list information, task configuration information, one or any combination thereof.

3. The method for managing federated learning tasks according to claim 1, wherein: The blockchain registers first local data of a client of a first tenant, and the first local data includes: one or any combination of data hash information, description information, and attribution information of the first tenant.

4. The method for managing federated learning tasks according to claim 1, wherein: Also includes: The management platform receives a second blockchain account registration request sent by the client of the second tenant; The management platform sends the information to be filled in to the client of the second tenant according to the second blockchain account registration request; The management platform receives second registration information fed back by the client of the second tenant according to the information to be filled in; The management platform initiates a second identity information binding request to the blockchain, where the second identity information binding request carries the second registration information.

5. The method for managing federated learning tasks according to claim 4, wherein: The second registration information includes: identity information, public key information and computing node information of the second tenant.

6. The method for managing federated learning tasks according to claim 4, wherein: The blockchain registers second local data of a client of a second tenant, and the second local data includes: one or any combination of data hash information, description information, and attribution information of the second tenant.

7. A method for managing federated learning tasks, characterized in that: include: The client of the first tenant initiates a federated learning task to the management platform, wherein the federated learning task carries task information; After the first tenant's client receives the federated learning task initiated by the management platform on the blockchain, it initiates the federated learning task to the corresponding second tenant's client according to the task information. The management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the first tenant's client; After the second tenant's client receives the federated learning task, the first tenant's client and the second tenant's client start joint modeling and send the federated learning task progress information to the blockchain; After completing the joint modeling, the client of the first tenant and the client of the second tenant send the federated learning task result information to the blockchain; the client of the first tenant and the client of the second tenant actively report the task progress to the blockchain until the modeling is completed; the management platform obtains the task progress and results in real time through blockchain notification; the tenant of the task initiator views the task progress and results through the management platform; Also includes: The first tenant's client sends a first blockchain account registration request to the management platform; The first tenant's client receives the information to be filled in sent by the management platform according to the first registered blockchain account request; The client of the first tenant sends the first registration information corresponding to the information to be filled in to the management platform, and the first registration information is used to be carried in the first identity information binding request initiated by the management platform to the blockchain; wherein each tenant accesses the management platform through a browser and fills in the registration information, which includes the identity information, public key information, and computing node information required to apply for a blockchain account.

8. The method for managing federated learning tasks according to claim 7, wherein: The task information includes: initiator information, participant information, data hash list information, task configuration information, one or any combination thereof.

9. The method for managing federated learning tasks according to claim 7, wherein: Also includes: The client of the first tenant registers first local data in the blockchain, where the first local data includes: one or any combination of data hash information, description information, and attribution information of the first tenant.

10. The method for managing federated learning tasks according to claim 7, wherein: Also includes: The second tenant's client sends a second blockchain account registration request to the management platform; The second tenant's client receives the information to be filled in sent by the management platform according to the second registered blockchain account request; The client of the second tenant sends the second registration information corresponding to the information to be filled in to the management platform, and the second registration information is used to be carried in the second identity information binding request initiated by the management platform to the blockchain.

11. The method for managing federated learning tasks according to claim 10, wherein: Also includes: The client of the second tenant registers second local data in the blockchain, where the second local data includes: one or any combination of data hash information, description information, and attribution information of the second tenant.

12. A management platform, characterized in that: include: A task receiving module, used to receive a federated learning task initiated by a client of the first tenant, wherein the federated learning task carries task information; A task initiation module, used to initiate the federated learning task to the blockchain; An information receiving module is used for the management platform to receive the federated learning task progress information and federated learning task result information fed back by the blockchain, wherein the federated learning task result information is sent to the blockchain by the client of the first tenant and the client of the second tenant after completing the joint modeling, and the federated learning task progress information is sent to the blockchain when the client of the first tenant and the client of the second tenant perform joint modeling, and the second tenant's client starts the joint modeling after receiving the federated learning task initiated by the client of the first tenant, and the first tenant's client initiates the federated learning task to the corresponding client of the second tenant according to the task information after receiving the federated learning task initiated by the management platform on the blockchain; wherein the client of the first tenant and the client of the second tenant actively report the task to the blockchain The management platform obtains the task progress and results in real time through blockchain notification; the tenant of the task initiator views the task progress and results through the management platform; the management platform receives the first blockchain account registration request sent by the client of the first tenant; the management platform sends the information to be filled in to the client of the first tenant according to the first blockchain account registration request; the management platform receives the first registration information fed back by the client of the first tenant according to the information to be filled in; wherein each tenant accesses the management platform through a browser and fills in the registration information, which includes the identity information, public key information, and computing node information required for applying for a blockchain account; the management platform initiates a first identity information binding request to the blockchain, and the first identity information binding request carries the first registration information.

13. A client, characterized in that: include: A first task initiating module, used for a client of a first tenant to initiate a federated learning task to a management platform, wherein the federated learning task carries task information; The second task initiation module is used for the client of the first tenant to initiate the federated learning task to the client of the corresponding second tenant according to the task information after the blockchain receives the federated learning task initiated by the management platform. The management platform initiates the federated learning task to the blockchain after receiving the federated learning task initiated by the client of the first tenant; A joint modeling module, configured to enable the first tenant's client and the second tenant's client to start joint modeling and send the progress information of the federated learning task to the blockchain after the second tenant's client receives the federated learning task; An information sending module is used for the client of the first tenant and the client of the second tenant to send federated learning task progress information to the blockchain when performing joint modeling, and to send federated learning task result information to the blockchain after completing the joint modeling; wherein the client of the first tenant and the client of the second tenant actively report the task progress to the blockchain until the modeling is completed; the management platform obtains the task progress and results in real time through blockchain notification; the tenant of the task initiator views the task progress and results through the management platform; the client of the first tenant sends a first blockchain account registration request to the management platform; the client of the first tenant receives the information to be filled in sent by the management platform according to the first blockchain account registration request; the client of the first tenant sends the first registration information corresponding to the information to be filled in to the management platform, and the first registration information is used to be carried in the first identity information binding request initiated by the management platform to the blockchain; wherein each tenant accesses the management platform through a browser and fills in the registration information, which includes the identity information, public key information, and computing node information required to apply for a blockchain account.

14. A federated learning task management system, characterized in that: include: A management platform and a client, wherein the management platform is the management platform described in claim 12, and the client is the client described in claim 13.

15. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 11 is implemented.

16. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 11 is implemented.

Images