Model Information Acquisition Method, Transmission Method, Device, Node, and Storage Medium

Abstract

This application discloses a model information acquisition method, a transmission method, an apparatus, a node, and a storage medium, and belongs to the technical field of communications. The model information acquisition method according to an embodiment of this application includes: a step in which a model training functional node determines a federated learning FL server-side node; a step in which the model training functional node transmits a first request message for triggering the federated learning of the FL server-side node so as to acquire a target model to the FL server-side node; and a step in which the model training functional node receives information on the target model transmitted by the FL server-side node.

Description

Technical Field

[0001] (Cross - reference to related applications) This application claims the priority of a Chinese patent application with application number 202210476336.X filed in China on April 29, 2022, and all of its contents are incorporated herein by reference.

[0002] This application belongs to the technical field of communications, and specifically relates to a method for obtaining model information, a transmission method, an apparatus, a node, and a storage medium.

Background Art

[0003] Currently, some communication systems (for example, the 5th - Generation (5G) mobile communication) have introduced artificial intelligence (AI) functions. Specifically, those that process network services with models can be introduced. However, currently, only the independent learning method of network nodes is supported for obtaining models. Therefore, in some scenarios (for example, when the training data is insufficient), the network nodes may not be able to obtain models through training, resulting in the problem that the model training performance of the network nodes is poor.

Summary of the Invention

[0004] Embodiments of this application provide a method for obtaining model information, a transmission method, an apparatus, a node, and a storage medium to solve the problem that the model training performance of network nodes is poor.

[0005] In a first aspect, a model training function node determines a Federated learning (FL) server - side node, and the model training function node sends a first request message for triggering the federated learning of the FL server - side node to obtain a target model to the FL server - side node. The method for acquiring model information includes the step of the model training function node receiving information on the target model transmitted by the FL server-side node.

[0006] In a second aspect, The step of the federated learning FL server-side node receiving a first request message transmitted by the model training function node, where the first request message is for triggering the federated learning of the FL server-side node to acquire a target model, The step of the FL server-side node performing federated learning with the FL client node based on the first request message and acquiring the target model, The method for transmitting model information includes the step of the FL server-side node transmitting information on the target model to the model training function node.

[0007] In a third aspect, A first determination module for determining a federated learning FL server-side node, A first transmission module for transmitting a first request message for triggering the federated learning of the FL server-side node to acquire a target model to the FL server-side node, A receiving module for receiving information on the target model transmitted by the FL server-side node, and a model information acquisition device is provided.

[0008] In a fourth aspect, A receiving module for receiving a first request message transmitted by a model training function node, where the first request message is for triggering the federated learning of a federated learning FL server-side node to acquire a target model, A learning module for performing federated learning with an FL client node based on the first request message and acquiring the target model, A model information transmission device comprising: a transmission module for transmitting information of the target model to the model training function node.

[0009] In a fifth aspect, a model training function node is provided, which includes a processor and a memory, and a program or command executable by the processor is stored in the memory. When the program or command is executed by the processor, the steps of the model information acquisition method provided in the embodiments of the present application are realized.

[0010] In a sixth aspect, a model training function node is provided, which includes a processor and a communication interface. The processor or the communication interface is used to determine the FL server-side node, and the communication interface is used to: send a first request message to the FL server-side node to trigger the federated learning of the FL server-side node to obtain a target model; and receive information of the target model sent by the FL server-side node.

[0011] In a seventh aspect, a server-side node is provided, which includes a processor and a memory, and a program or command executable by the processor is stored in the memory. When the program or command is executed by the processor, the steps of the model information transmission method provided in the embodiments of the present application are realized.

[0012] In an eighth aspect, a server - side node is provided that includes a processor and a communication interface. The communication interface is used to receive a first request message sent by a model training function node. The first request message is for triggering the federated learning of a federated learning FL server - side node to obtain a target model. The processor or the communication interface is used to perform federated learning with an FL client node based on the first request message to obtain the target model. The communication interface is used to send information about the target model to the model training function node.

[0013] In a ninth aspect, a model information transmission system is provided that includes a model training function node and a server - side node. The model training function node can be used to execute the steps of the model information acquisition method provided in the embodiments of the present application. The server - side node can be used to execute the steps of the model information transmission method provided in the embodiments of the present application.

[0014] In a tenth aspect, a readable storage medium is provided in which a program or command is stored. When the program or command is executed by a processor, the steps of the model information acquisition method provided in the embodiments of the present application are realized, or the steps of the model information transmission method provided in the embodiments of the present application are realized.

[0015] In an eleventh aspect, a chip is provided that includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is for executing a program or command to realize the model information acquisition method provided in the embodiments of the present application, or to realize the model information transmission method provided in the embodiments of the present application.

[0016] In a 12th aspect, there is provided a computer program / program product that is stored in a storage medium and, when executed by at least one processor, implements the steps of the model information acquisition method provided in an embodiment of the present application or implements the model information transmission method provided in an embodiment of the present application.

[0017] In an embodiment of the present application, a model training function node determines an FL server-side node, and the model training function node sends a first request message to the FL server-side node for triggering the federated learning of the FL server-side node to obtain a target model, and the model training function node receives information of the target model sent by the FL server-side node. In this way, the model training function node can obtain information of the target model through federated learning and improve the model training performance of network nodes.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Best Mode for Carrying Out the Invention

[0019] In the following, while referring to the drawings in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly described. Naturally, the described embodiments are part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art shall fall within the protection scope of the present application.

[0020] The terms "first", "second", etc. in the specification and claims of the present application are not for describing a specific order or sequence, but for distinguishing similar objects. It should be understood that such terms may be replaced with each other in appropriate cases so that the embodiments of the present application can be implemented in an order other than that illustrated or described herein. Moreover, the objects distinguished by "first" and "second" are usually of one kind, and the number of objects is not limited. For example, the first object may be one or a plurality. Also, in the specification and claims, "and / or" represents at least one of the connected objects, and the symbol " / " generally represents that the related objects before and after are in an "or" relationship.

[0021] It should be noted that the technology described in the embodiments of this application is not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems. For example, it can also be used in other wireless communication systems such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency Division Multiple Access (SC-FDMA), and other systems. In the embodiments of this application, the terms "system" and "network" are often used interchangeably, and the technology described can be used in other systems and radio communication technologies in addition to the systems and radio communication technologies described above. In the following description, a New Radio (NR) system is described for illustrative purposes, and the NR term is used in many of the following descriptions. However, these technologies are also applicable to applications other than NR system applications, such as 6th Generation (6G) communication systems.

[0022] FIG. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 may be a mobile phone, a tablet personal computer (TPC), a laptop computer (LC) also called a notebook computer, a personal digital assistant (PDA), a portable information terminal, a netbook, an ultra-mobile personal computer (UMPC), a mobile internet device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, a vehicle user equipment (VUE), a pedestrian user equipment (PUE), a smart home (home devices having a wireless communication function such as a refrigerator, a television, a washing machine, or furniture), a game console, a personal computer (PC), a cash dispenser, or a kiosk. The wearable device may include a smart watch, a smart wristband, smart earphones, smart glasses, smart jewelry (smart bangle, smart bracelet, smart ring, smart necklace, smart anklet bangle, smart anklet, etc.), a smart list strap, a smart wear, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiments of the present application.

[0023] The network-side device 12 can include an access network device and a core network device. The access network device may be referred to as a radio access network device, a radio access network (RAN), a radio access network function, or a radio access network unit. The access network device may include a base station, a wireless local area network (WLAN) access point, or a wireless fidelity (WiFi) node, etc. The base station may be referred to as Node B, evolved Node B (eNB), access point, base transceiver station (BTS), radio base station, radio transceiver, basic service set (BSS), extended service set (ESS), home Node B, home evolved B node, transmitting receiving point (TRP), or other suitable terms in the field. If the same technical effect can be achieved, the base station is not limited to a specific technical term. It should be noted that in the embodiments of this application, only the base station in the NR system is described as an example, and the specific type of the base station is not limited.

[0024] The core network device may include, but is not limited to, at least one of a core network node, a core network function, a Mobility Management Entity (MME), an Access and Mobility Management Function (AMF), a Session Management Function (SMF), a User Plane Function (UPF), a Policy Control Function (PCF), a Policy and Charging Rules Function unit (PCRF), an Edge Application Server Discovery Function (EASDF), a Unified Data Management (UDM), a Unified Data Repository (UDR), a Home Subscriber Server (HSS), a Centralized network configuration (CNC), a Network Repository Function (NRF), a Network Exposure Function (NEF), a Local NEF (or L-NEF), a Binding Support Function (BSF), an Application Function (AF), a Network Data Analytics Function (NWDAF), etc. It should be noted that in the embodiments of this application, only the core network device in the NR system is taken as an example for description, and the specific types of the core network device are not limited.

[0025] In an embodiment of the present application, the model training function node may be a network node for generating a model and performing model training, and may refer to a general model training function node (not an FL server-side node) or an FL client node. That is, the fact that the model training function node is not an FL server-side node means that the model training function node cannot be used as an FL server-side node for a specific model training task (corresponding to a specific analysis identifier, a specific area of interest (AOI)), while it can be used as an FL server-side node for other model training tasks (for example, other analysis identifiers, other AOIs), but it should be understood that it is not limited to this case.

[0026] The model inference function node may be a network node for performing inference and generation of prediction information, generation of statistical information, data analysis, etc.

[0027] In some embodiments, the model training function node may refer to a network element, a terminal, or a module with an AI model training function in a communication network, and the model inference function node refers to a network element, a terminal, or a module with a model inference function in a communication network. That is, the model training function node and the model inference function node may of course be called by other names.

[0028] In the embodiments of the present application, the model training function node and the model inference function node may be core network elements or modules within the core network element. For example, NWDAF may include a model training function node and a model inference function node. The FL server-side node is a core network element with the function of a federated learning server, and the FL client node may be another core network element or module participating in the federated learning. Or, the model training function node and the model inference function node may be radio access network elements or modules within the radio access network element, that is, functions within the RAN. For example, the model training function node may be a RAN device with the model training function. Specifically, for example, it may be a base station device or module with the model training function but not an FL server-side node. The FL server-side node is a base station device or module with the function of a federated learning server or operation administration and maintenance (OAM). The FL client node may be another base station device or module that becomes a member participating in the federated learning.

[0029] In addition, the model training function node and the model inference function node may be functions of a terminal or within the terminal. For example, the model training function node may be a terminal with the model training function. Specifically, for example, it may be a terminal with the model training function but not an FL server-side node. The FL server-side node is a terminal or AF with the function of a federated learning server, and the other FL client nodes may be other terminals that become members participating in the federated learning.

[0030] It should be noted that in the embodiments of the present application, the model training function node and the model inference function node may be arranged as independent and different network element devices, or may be jointly arranged in the same network element device, such as a core network element, a radio access network element, or two functional modules inside a terminal. In this case, the core network element, the radio access network element, or the terminal can provide not only the AI model training function but also the model inference function.

[0031] In the embodiments of the present application, the FL server-side node is a network element that coordinates or controls federated learning. For example, it may be an FL center network element or an FL coordinator network element, or a central node of the model training function that can be used for FL operations. Specifically, it may be a network element such as a core network element, a radio access network element, a terminal, or an application server. The FL client node is a network element that participates in federated learning and may be called an FL operation participation network element. Specifically, it may also be a network element such as a core network element, a radio access network element, a terminal, or an application server.

[0032] In some embodiments, the model training function node may be a Model Training Logical Function (MTLF), the model inference function node may be an Analytics Logical Function (AnLF), the FL server-side node may be an FL server, and the FL client node may be an FL client.

[0033] Referring to FIG. 2, which is a flowchart of the model information acquisition method provided in the embodiments of the present application, as shown in FIG. 2, the method includes the following steps 201 to 203.

[0034] In step 201, the model training function node determines the FL server-side node.

[0035] In the above step of determining the FL server-side node, the model training function node may query the above FL server-side node from other network nodes, or select the above FL server-side node according to the pre-acquired configuration information based on the model training function node.

[0036] In step 202, the model training function node sends a first request message to the FL server-side node to trigger the federated learning of the FL server-side node to obtain the target model.

[0037] With the above first request message, the FL server-side node can be requested to perform federated learning and training to obtain the above target model. Specifically, the FL server may execute FL operations with at least one FL client node. For example, the FL server may perform an interactive iterative process of federated learning with at least one FL client node to obtain the above target model. Since the above target model is a model obtained through federated learning, it may also be called a federated learning model.

[0038] In step 203, the model training function node receives the information of the target model sent by the FL server-side node.

[0039] In this step, after the FL server-side node obtains the above target model through federated learning, the information of the target model may be sent to the model training function node.

[0040] The above information of the target model may be information for determining the above target model, such as model file information of the target model, or download address information of the model file.

[0041] It should be noted that in the embodiments of the present application, the above target model may be a model for processing communication services in a communication system. For example, it may be a model for data analysis, or a model for inference tasks, or a model for channel estimation, or a model for information prediction, etc.

[0042] In the embodiments of the present application, through the above steps, the model training function node can obtain the information of the target model through federated learning, thereby improving the model training performance of the network node and further solving the problem that the target model cannot be obtained due to data privacy reasons. For example, when the model training function node needs to provide a target model for the Area of Interest (AOI) specified by the model inference function node, but it is found that all or part of the training data within the AOI cannot be obtained due to the problem of data islands, the federated learning of the FL server-side node is triggered.

[0043] As an alternative embodiment, the step in which the model training function node determines the FL server-side node includes: The step in which the model training function node sends a node discovery request message for requesting network nodes participating in federated learning training to the network repository function network element; The step in which the model training function node receives a response message sent by the network repository function network element, where the response message includes information of the FL server-side node.

[0044] Here, the above node discovery request message may be a Network Element Discovery Request (Nnrf_NFDiscovery_Request).

[0045] Since the network repository function network element can store information of one or more FL server-side nodes, after receiving the request message, it returns the information of the corresponding FL server-side node to the model training function node.

[0046] In this embodiment, it is possible to realize the acquisition of information of the FL server-side node by the network repository function network element.

[0047] It should be noted that in some embodiments, it is not necessary to use the network repository function network element to acquire information of the FL server-side node. For example, the information of the FL server-side node may be fixedly set in the model training function node.

[0048] Optionally, the node discovery request message includes at least one of an analytics ID, area of interest (AOI) information, area of interest time information, model description method information, model sharable information, model performance information, model algorithm information, model training speed information, federated learning instruction information, federated learning type information, FL server-side node type instruction information, FL client node type instruction information, first service information, and second service information.

[0049] The analytics ID can be used to indicate that the network node requested by the request message should support the model training task corresponding to the analytics ID.

[0050] The AOI information can be used to indicate that the network node requested by the request message can provide services in at least one tracking area (TA), at least one cell, or other areas that may be areas corresponding to the AOI information.

[0051] The above-mentioned time of interest information can be used to instruct that the network node requested by the request message should support model training for the time period corresponding to the time of interest information.

[0052] The above-mentioned model description method information, which may also be referred to as model description method requirement information or model description method expectation information, can be used to instruct that the network node requested by the request message should support a model representation based on the model description method corresponding to the model description method information. For example, the model description method information can specifically be a model representation language represented by Open Neural Network Exchange (ONNX) or the like, or a model framework represented by TensorFlow, Pytorch, or the like.

[0053] The above-mentioned model sharability information, which may also be referred to as model sharability requirement information or model sharability expectation information, can be used to instruct that the network node requested by the request message needs to be able to share the model with the model training functional node. Here, "sharable" means being interoperable, or being understandable and executable with each other.

[0054] The above-mentioned model performance information, which may also be referred to as model performance requirement information or model performance expectation information, can be used to instruct that the network node requested by the request message needs to provide a model that can meet the model performance information. Here, the performance can be the accuracy value of the model, the Mean Absolute Error (MAE), or the like.

[0055] The above model algorithm information, which may also be referred to as model algorithm requirement information or model algorithm expectation information, can be used to instruct that the network node requested by the request message should support the training of a model based on the model algorithm corresponding to the model algorithm information.

[0056] The above model training speed information, which may also be referred to as model training speed requirement information or model training speed expectation information, can be used to instruct that the speed of model training by the network node requested by the request message should meet the model training speed indicated by the model training speed information. Here, the training speed may be expressed as the time taken for the model training to converge or reach a certain performance threshold.

[0057] The above federated learning instruction information can be used to instruct that the network node requested by the request message should support federated learning.

[0058] The above federated learning type information is can be used to instruct that the type of federated learning that the network node requested by the request message should support is at least one of horizontal federated learning and vertical federated learning.

[0059] Here, the type of the above horizontal federated learning may be one that performs learning training with different training data samples having the same feature points, and the type of the above vertical federated learning may be one that performs learning training with training data samples of different feature points of the same training sample.

[0060] The above FL server-side node type instruction information can be used to instruct that the network node requested by the request message belongs to the FL server-side node type.

[0061] The above FL client node type indication information can be used to indicate that the network node requested by the above request message belongs to the FL client node type.

[0062] The above first service information can be used to indicate that the network node requested by the above request message should support the service of the federated learning server.

[0063] The above second service information can be used to indicate that the network node requested by the above request message should support the service of the federated learning member.

[0064] In this embodiment, since the above node discovery request message includes the above at least one piece of information, the network repository function network element can return the information of the FL server-side node that meets the corresponding conditions to the model training function node, so that the FL server-side node can finally obtain the required, acceptable or desired target model of the above model training function node through federated learning.

[0065] Optionally, the response message includes information of N network nodes including the above FL server-side node, where N is a positive integer, The information of each network node is including at least one of a Fully Qualified Domain Name (FQDN), identifier information, and address information.

[0066] In this embodiment, the above FL server-side node is determined by at least one of the above FQDN, identifier information, and address information.

[0067] In some embodiments, the above N network nodes further include FL client nodes.

[0068] One or more FL client nodes included in the above N network nodes may be FL client nodes participating in the current federated learning. These FL client nodes can be FL participants or FL members.

[0069] Since the response message further includes FL client nodes, the model training function node can quickly determine the FL server-side nodes and FL client nodes, and thus, by timely requesting the FL server-side nodes and FL client nodes to perform federated learning, the acquisition efficiency of the target model can be improved.

[0070] Optionally, the information of each network node further includes type information for indicating the type of the network node, and the type is one of the FL server-side node and the FL client node.

[0071] In this embodiment, the type information can indicate that the network node is an FL server-side node and an FL client node. Of course, in the embodiments of the present application, the above type information may not be indicated. For example, the model training function node can identify whether it is an FL server-side node or an FL client node based on the identifier information of these network nodes, or when the above response message includes the FL server-side node and the FL client node, in the response message, the information of the FL server-side node and the FL client node is sorted, and thus, the model training function node can identify the FL server-side node and the FL client node by the sorting of the information.

[0072] As an alternative embodiment, the first request message includes at least one of federated learning instruction information and a model identifier, wherein the federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model. The model identifier is for uniquely identifying the target model.

[0073] Here, the model identifier may be obtained by the model training function node. For example, the method may further include the step of the model training function node obtaining the model identifier, such as by self-generation or by obtaining it from a model identifier management network element.

[0074] In this embodiment, since the first request message includes at least one of the federated learning instruction information and the model identifier, the FL server-side node can quickly respond to the request message by determining to perform federated learning. It should be noted that in some embodiments, the federated learning instruction information and the model identifier may not be carried. For example, the model identifier may be obtained by the FL server-side node, whereby the FL server-side node will perform federated learning by default when receiving the first request message.

[0075] In some embodiments, the first request message may include at least one of an analysis identifier, time-of-interest information, model description method information, model sharability information, model performance information, model algorithm information, model training speed information, first model filter information, first target information, and reporting information. For the analysis identifier, time-of-interest information, model description method information, model sharability information, model performance information, model algorithm information, and model training speed information, reference may be made to the corresponding descriptions in the above embodiments, and detailed descriptions are omitted here.

[0076] The first model filter information can be used to indicate at least one of the area range, time range, slice range, and data network name (DNN) range of federated learning.

[0077] The first target information can be used to instruct that the federated learning targets one of a single terminal, multiple terminals, or any terminal. For example, the first target information includes terminal information such as a terminal identifier (UE ID) or a terminal group identifier (UE group ID).

[0078] The report information can be used to instruct the transmission of the report format and / or report conditions of the federated model.

[0079] In this embodiment, since the first request message includes the at least one piece of information, the FL server-side node can finally obtain the required, acceptable, or desired target model of the model training functional node through federated learning.

[0080] As an alternative embodiment, the first request message includes information about FL client nodes participating in the federated learning.

[0081] The FL client nodes participating in the federated learning may be one or more FL client nodes, and these clients may or may not include the model training functional node.

[0082] The information about the FL client nodes obtained through the federated learning may be obtained from a network repository function network element, or may be preset by the model training functional node. Since the first request message includes the information about the FL client nodes participating in the federated learning, the FL server-side node can improve the acquisition efficiency of the target model by quickly performing federated learning with the FL client nodes.

[0083] As an alternative embodiment, the information about the target model is It includes at least one piece of information among a model identifier, federation instruction information, a model file, and address information of the model file corresponding to the target model.

[0084] The above collaborative learning instruction information is for instructing that the target model is a model obtained by collaborative learning.

[0085] The model identifier is for uniquely identifying the target model.

[0086] The above model file may include information on related files of the above target model, such as files of the network structure, weight parameters, and input / output data of the model.

[0087] The address information of the above model file is address information for obtaining the above target model, such as a storage address for instructing the model file or download address information of the model file.

[0088] It should be noted that in the embodiments of the present application, since the information of the above target model can be self-acquired by the model training function node for the identifier, the above model identifier may not be included. Since the model training function node can, by default, recognize that the target model transmitted by the FL server-side node is a model obtained by collaborative learning, the above collaborative instruction information may not be included. Since the model training function node downloads the model file based on the above address information, the above model file may not be included. Also, in some embodiments, after the model is trained, the model file can be stored at a preset position. Therefore, since the model training function node can download the model file from this position, the address information of the model file may not be included.

[0089] In some embodiments, the information of the above target model may further include at least one piece of information among an analysis identifier, second model filter information, and model validity information corresponding to the above target model.

[0090] The analysis identifier can be used to indicate a task corresponding to the target model.

[0091] The second model filter information can be used to indicate at least one of the area range, time range, slice range, and data network name (DNN) range of the target model.

[0092] The model validity information can be used to indicate validity information such as the valid time and valid area of the model.

[0093] As an alternative embodiment, the method further includes a step in which the model training function node transmits information of the target model to the model inference function node.

[0094] In this embodiment, the information of the target model may be spontaneously transmitted to the model inference function node, or the information of the target model may be transmitted based on a request from the model inference function node.

[0095] For example, in the form of transmission by request, the model training function node receives a model request message transmitted by the model inference function node. The model request message may include at least one of an analysis identifier, third model filter information, second target information, and time information, and the analysis identifier can be used to indicate a data analysis task targeted by the model identifier requested by the model request message.

[0096] The above-mentioned third model filter information can be used to indicate the conditions that the model required by the model request message should meet. For example, the conditions may be the area of interest where the model needs to provide services, single network slice selection assistance information (S-NSSAI), or DNN.

[0097] The above-mentioned second target information can be used to indicate the training target of the model required by the model request message. The training target may be a single terminal, multiple terminals, or any terminal, etc. For example, the second target information may be the terminal identifier or terminal group identifier of these terminals.

[0098] The time information is for indicating at least one of the application time of the model required by the model request message and the reporting time of the model information.

[0099] In this embodiment, the model training function node transmits the information of the target model to the model inference function node. Therefore, the model inference function node can make the above target model available to improve service performance.

[0100] As an alternative embodiment, before the step where the model training function node determines the FL server-side node, the method further includes the step of the model training function node determining that the target model should be obtained by federated learning.

[0101] The above step of determining that the target model should be obtained by federated learning may be that the model training function node determines that the target model should be obtained by federated learning because the target model cannot be obtained by independent training. For example, the above step of the model training function node determining that the target model should be obtained by federated learning may be When the model training function node determines that it cannot obtain all or part of the training data for generating the target model, it includes the step of determining that the target model should be obtained by federated learning.

[0102] The inability to obtain all or part of the training data for generating the target model described above may be due to reasons such as technology confidentiality or user privacy, where the model training function node cannot obtain all or part of the training data for generating the target model. For example, when the model training function node needs to provide a target model for the AOI specified by the model inference function node, but discovers that it may not be able to obtain all or part of the training data within the AOI due to the data island problem, it triggers the federated learning of the FL server side node. Also, for example, when the model training function node needs to obtain a model corresponding to some terminals, but cannot obtain the training data corresponding to these terminals due to data privacy issues, it may rely on the FL server side node to obtain the training data corresponding to these terminals, or trigger federated learning to obtain the target model for the terminals.

[0103] In this embodiment, when it is determined that the target model should be obtained by federated learning, by obtaining the information of the target model by the FL server side node, the model training performance of the network node can be improved. For example, even when the model training function node does not have sufficient training data for the target model, it can obtain the target model, and even when it does not have sufficient training resources for the target model, it can obtain the target model, thereby improving the model training performance of the network node.

[0104] In the embodiments of the present application, the model training function node determines the FL server-side node, and the model training function node sends a first request message for triggering the federated learning of the FL server-side node to obtain a target model to the FL server-side node, and the model training function node receives the information of the target model sent by the FL server-side node. In this way, the model training function node can obtain the information of the target model through federated learning, thereby improving the model training performance of the network node.

[0105] Referring to FIG. 3 which is a flowchart of the model information transmission method provided in the embodiments of the application, as shown in FIG. 3, the method includes the following steps 301 to 303.

[0106] In step 301, the FL server-side node receives the first request message sent by the model training function node, and the first request message is for triggering the federated learning of the FL server-side node to obtain a target model.

[0107] In step 302, the FL server-side node performs federated learning with the FL client node based on the first request message to obtain the target model.

[0108] Here, the above step in which the FL server-side node performs federated learning with the FL client node based on the first request message may trigger the execution of the FL operation between the FL server-side node and the FL client node by the first request message. Specifically, the target model may be trained and obtained by the FL server-side node and the FL client node performing an interactive iterative process of federated learning.

[0109] In step 303, the FL server-side node sends the information of the target model to the model training function node.

[0110] Optionally, the first request message includes at least one of federated learning instruction information and a model identifier, wherein the federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model, and the model identifier is for uniquely identifying the target model.

[0111] In some embodiments, the first request message may include at least one of an analysis identifier, interest time information, model description method information, model sharable information, model performance information, model algorithm information, model training speed information, first model filter information, first target information, and report information.

[0112] Optionally, the first request message includes information about FL client nodes participating in federated learning.

[0113] Optionally, the method further includes a step of the FL server-side node determining FL client nodes participating in federated learning.

[0114] Here, the step of the FL server-side node determining FL client nodes participating in federated learning may be that the FL server-side node queries the FL client nodes participating in federated learning from a network repository function network element, or the FL server-side node determines the FL client nodes participating in federated learning based on preset information.

[0115] Optionally, the step of the FL server-side node determining FL client nodes participating in federated learning includes the step of the FL server-side node sending a node discovery request message for requesting FL client nodes participating in federated learning to a network repository function network element, The step in which the FL server - side node receives a response message sent by the network repository function network element, wherein the response message includes information on FL client nodes participating in federated learning, is included.

[0116] Optionally, the information of the federated model includes at least one piece of information among a model identifier, federation instruction information, a model file, and address information of the model file corresponding to the target model. The federated learning instruction information is for instructing that the target model is a model obtained by federated learning. The model identifier is for uniquely identifying the target model.

[0117] In some embodiments, the information of the above - mentioned target model may further include at least one piece of information among an analysis identifier, second model filter information, and model validity information corresponding to the target model.

[0118] Optionally, the model identifier is obtained by the model training function node for the target model.

[0119] Optionally, the method further includes the step in which the FL server - side node obtains the model identifier for the target model.

[0120] Here, the model identifier obtained by the FL server - side node for the target model may be generated by the FL server - side node or obtained from a model identifier management network element.

[0121] It should be noted that this embodiment corresponds to the embodiment of the FL server - side node shown in FIG. 2. For its specific implementation, reference may be made to the relevant description of the embodiment shown in FIG. 2. To avoid repeated description, detailed description is omitted in this embodiment.

[0122] In the following, taking the model training function node as MTLF, the model inference function node as AnLF, the FL server - side node as the FL server, and the FL client node as the FL client as an example, the method provided in the embodiments of this application will be described by means of a plurality of embodiments.

[0123] Embodiment 1 In this embodiment, the determination of the FL client by the FL server will be described by way of example. Specifically, as shown in FIG. 4, it includes the following steps 0 - 2, steps 3a - 3b, steps 4 - step 7, step 7a, steps 8 - 11.

[0124] In step 0, optionally, the Consumer Network Function (Consumer NF) requests the data analysis result from AnLF, and the request message can carry at least one of the analysis identifier, analysis filter information, target of analytic reporting, and analytics reporting information.

[0125] The analytics identifier is for identifying the type of data analysis task. For example, analytics ID = UE mobility is for indicating a task that requests user movement trajectory data analysis.

[0126] The analytic filter information is for indicating data analysis result filter information including the Area of Interest (AOI), slice S - NSSAI, DNN, etc.

[0127] The target of analytic reporting is for explicitly indicating that the data analysis target is a certain terminal, multiple terminals, or all UEs.

[0128] The analytics reporting information is for indicating information such as the reporting time and reporting conditions of the data analysis result.

[0129] In step 1, AnLF sends a model request message to MTLF1 to request a model corresponding to the analysis identifier.

[0130] The model request message may be a model provision subscription (Nnwdaf_MLModelProvision_subscribe) message or a model information request message (Nnwdaf_MLModelInfo_Request).

[0131] The model request message may carry Analytic ID, model filter information, model target information, and model time information.

[0132] Analytic ID is for identifying the type of inference task. For example, analytics ID = UE mobility is for predicting the user's movement trajectory.

[0133] Model filter information is for indicating conditions that the requested model should meet, such as area of interest (AOI), slice S-NSSAI, DNN.

[0134] Model target information is for indicating the model training target, such as a single terminal, multiple terminals, or any terminal.

[0135] Model time information is for indicating the application time of the model or the reporting time of the model information.

[0136] It should be noted that MTLF1 here is not the FL server, specifically not the FL server of the target model, but can be the FL server of other models, but is not limited thereto. The FL server may be called a coordinator or a central MTLF for FL operation, but it is just one FL client (or a member participating in the FL operation) or an MTLF that does not support FL capabilities.

[0137] In step 2, MTLF1 needs to determine to perform federated learning to obtain the required model based on the model request message.

[0138] The factors for MTLF1 to determine that federated learning needs to be performed may include the following.

[0139] In MTLF1, there is not enough training data for the model training task. For example, due to data privacy issues, specific training data in other regions or other training entities is lacking, and centralized model training for generating the required model cannot be performed.

[0140] Optionally, MTLF1 can also determine that the type of federated learning is horizontal federated learning. For example, MTLF1 determines that horizontal federated learning needs to be performed based on the fact that the training task satisfies the feature that the samples corresponding to the training data are different but the features are the same.

[0141] In step 3a, optionally, MTLF1 sends a node discovery request message for requesting network element devices capable of performing federated learning training to the NRF.

[0142] Specifically, as the node discovery request message, Discovery Request (Nnrf_NFDiscovery_Request) can be used. Here, the message may include an analytics identifier and an AOI.

[0143] The analytics identifier is for indicating that the requested network element should support the analytics ID.

[0144] The AOI may be at least one Tracking Area (TA), at least one cell, or other forms, and is for indicating that the requested network element needs to provide services for the AOI.

[0145] The node discovery request message may further include at least one of the following: federated learning indication information, information that the expected NF type is an FL server, information that the expected NF type is an FL client, information that the expected NF service name is a collaborative service FL service, and information that the expected NF service name is a service participation FL client.

[0146] The federated learning indication information is for indicating that the requested network element needs to support federated learning. Furthermore, it can also be used to indicate that horizontal federated learning should be supported.

[0147] The information that the expected NF type is an FL server (Expected NF type = FL server) is for indicating that the requested network element should belong to the FL server type.

[0148] The information that the expected NF type is an FL client (Expected NF type = FL client) is for indicating that the required network element should belong to the FL client type.

[0149] The information that the expected NF service name is the FL server (coordination) service (Expected NF service Name = FL server(coordination)service) is for indicating that the required network element should support the service of the federated learning server (coordination).

[0150] The information that the expected NF service name is the FL client (participant) service (Expected NF service Name = FL client(particepant)service) is for indicating that the required network element should support the service of the federated learning member (participant).

[0151] In one implementation, the network element request message that can specify the analysis identifier, the expected NF type is MTLF (Expected NF type = MTLF), and the AOI includes federated learning instruction information. In this case, MTLF1 is for requesting from the NRF all MTLFs that support federated learning for the AOI.

[0152] In one implementation, the network element request message includes the information that the expected NF type is an FL server (Expected NF type = FL server), and the rest is the same as related technologies (for example, analytics ID, Expected NF type = MTLF, specification of AOI). In this case, MTLF1 is for requesting from the NRF the FL servers that support federated learning.

[0153] In one implementation, the network element request message includes the information "Expected NF type=FL server and FL client", and the rest is the same as related technologies (e.g., analytics ID, Expected NF type=MTLF, specification of AOI). In this case, MTLF1 is for requesting the NRF for FL servers and FL clients that support federated learning.

[0154] In step 3b, the NRF returns to MTLF1 the device information that satisfies the request message initiated by the network element in step 3a.

[0155] The NRF sends to MTLF1 the FQDN, identifier information, address information, etc. of one or more MTLFs that satisfy the request, and the feedback information can explicitly indicate whether it is an FL server or an FL client for each MTLF.

[0156] In step 4, optionally, MTLF1 assigns a model identifier (model ID) to the federated learning model to be generated to uniquely identify the model.

[0157] In step 5, MTLF1 sends a federated learning request message to the FL server to request the FL server to trigger the federated learning process. Here, the federated learning request message may include at least one of the information of federated learning instruction, analytics identifier, model identifier, model filter information, model target, and model report information.

[0158] The federated learning instruction (FL indication) is for instructing to request to perform the federated learning process.

[0159] The analytics identifier (Analytics ID) is for instructing to request the task type for the analytics ID identifier and perform the federated learning process.

[0160] The model identifier (Model ID) is for uniquely identifying the model by federated learning.

[0161] The model filter information is for limiting the scope of the federated learning process, such as the area range, time range, S-NSSAI, DNN, etc.

[0162] The model target of the model is for specifying the target of the federated learning process, such as one or more specific terminals, all terminals, etc.

[0163] The model reporting information is for instructing the reporting information of the generated federated learning model information, such as the reporting time (start time, end time, etc.), reporting conditions (periodic trigger, event trigger, etc.).

[0164] In step 6, optionally, if the model identifier has not been received from MTLF1, the FL server may assign a model identifier for uniquely identifying the model to the federated learning model to be generated.

[0165] In step 7a, the FL server determines at least one FL client for performing the current federated learning process. Specifically, the FL server can query from the network repository function network element to obtain the FL client that meets the current federated learning process. Refer to step 3a.

[0166] It should be noted that the FL client here may or may not include MTLF1 itself.

[0167] In step 7, the FL server and the FL client perform an interactive iterative process of federated learning to obtain a federated learning model.

[0168] Here, if the FL client does not include MTLF1 itself, MTLF1 does not participate in the interactive process here.

[0169] In step 8, the FL server sends the information of the target model of the generated federated learning model to MTLF1. Here, the information of the target model includes a model identifier, a federated learning instruction, a model file (including the network structure, weight parameters, input / output data, etc. of the model), model file download address information or storage address information (for indicating the storage address of the model file or the download address of the model file), an analysis identifier (for indicating that the model is applicable to a specific inference task type), model filter information (such as area range, time range, S-NSSAI, DNN, etc., for limiting the scope of the federated learning process), valid area information (the area to which the model is applicable), valid time information (the time to which the model is applicable), and includes at least one of them.

[0170] In step 9, MTLF1 sends the model information of the generated federated learning model to AnLF.

[0171] In this step, MTLF1 can send the model by means of a model provision notification (Nnwdaf_MLModelProvision_Notify) or a model information response (Nnwdaf_MLModelInfo_Response) message. For the specific content of the transmission, refer to step 8 specifically.

[0172] In step 10, AnLF generates a data analysis result based on the model.

[0173] In step 11, optionally, AnLF sends the data analysis result to the consumer NF.

[0174] Example 2 In this example, the case where MTLF1 determines the FL client is taken as an example for explanation. Specifically, as shown in FIG. 5, there are the following differences from Example 1.

[0175] In step 3, MTLF1 obtains the FL server and the FL client from the NRF.

[0176] In step 5, optionally, MTLF1 simultaneously indicates to the FL server the FL clients participating in the current federated learning. The step where the FL server queries the NRF for the FL clients may be omitted.

[0177] In the embodiments of the present application, when it is necessary to rely on federated learning to generate a model due to data privacy issues, even if the model training function node itself does not support the federated learning ability or the federated learning server function, or does not support the federated learning or the federated learning server function for a specific analysis identifier or a specific AOI, the model training function node can trigger other devices to perform federated learning and obtain the model it needs. Thereby, the application scope of federated learning is expanded, and the problems on data privacy are solved in a wider range.

[0178] Referring to FIG. 6 which is a structural diagram of the model information acquisition device provided in the embodiments of the present application, as shown in FIG. 6, the model information acquisition device 600 includes a first determination module 601 for determining the federated learning FL server side node, a first transmission module 602 for transmitting a first request message for triggering the federated learning of the FL server side node to obtain the target model to the FL server side node, a reception module 603 for receiving the information of the target model transmitted by the FL server side node.

[0179] Optionally, the first decision module 601 is used to send a node discovery request message to the network repository function network element to request network nodes participating in the federated learning training, and to receive the response message sent by the network repository function network element, where the response message includes information about the FL server-side nodes.

[0180] Optionally, the node discovery request message includes at least one of an analysis identifier, area of interest (AOI) information, time of interest information, model description method information, model sharable information, model performance information, model algorithm information, model training speed information, federated learning instruction information, federated learning type information, FL server-side node type instruction information, FL client node type instruction information, first service information, and second service information. The federated learning instruction information is for instructing that the network nodes requested by the request message should support federated learning. The first service information is for instructing that the network nodes requested by the request message should support the service of the federated learning server. The second service information is for instructing that the network nodes requested by the request message should support the service of the federated learning member.

[0181] Optionally, the federated learning type information indicates that the type of federated learning that the network nodes requested by the request message should support is horizontal federated learning and / or vertical federated learning.

[0182] Optionally, the response message includes information about N network nodes including the FL server-side nodes, where N is a positive integer.

[0183] The information of each network node includes at least one of a fully qualified domain name FQDN, identifier information, and address information.

[0184] Optionally, the N network nodes further include FL client nodes.

[0185] Optionally, the information of each network node further includes type information for indicating the type of the network node, and the type is one of an FL server-side node and an FL client node.

[0186] Optionally, the first request message includes at least one of federated learning instruction information and a model identifier, wherein the federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model, and the model identifier is for uniquely identifying the target model.

[0187] Optionally, the model information acquisition device 600 further includes an acquisition module for acquiring the model identifier.

[0188] Optionally, the first request message includes information of FL client nodes participating in federated learning.

[0189] Optionally, the information of the target model includes at least one of information of a model identifier, federated instruction information, a model file, and address information of the model file corresponding to the target model, wherein the federated learning instruction information is for indicating that the target model is a model obtained by federated learning, and the model identifier is for uniquely identifying the target model.

[0190] Optionally, the model information acquisition device 600 further includes a second transmission module for transmitting the information of the target model to the model inference function node.

[0191] Optionally, the model information acquisition device 600 further includes a second determination module for determining that the target model should be acquired by federated learning.

[0192] Optionally, the second determination module is used to determine that the target model should be acquired by federated learning when it is determined that the model training function node cannot obtain all or part of the training data for generating the target model.

[0193] The above model information acquisition device can improve the model training performance of network nodes.

[0194] The model information acquisition device in the embodiments of the present application may be an electronic device, such as an electronic device equipped with an operating system, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device may be a core network device, a network-side device or a terminal, or other devices other than the terminal.

[0195] Referring to FIG. 7 which is a structural diagram of the model information transmission device provided in the embodiments of the present application, as shown in FIG. 7, the model information transmission device 700

[0196] Referring to FIG. 7 which is a structural diagram of the model information transmission device provided in the embodiments of the present application, as shown in FIG. 7, the model information transmission device 700 A receiving module 701 for receiving a first request message sent by a model training function node, wherein the first request message is for triggering federated learning of a federated learning FL server-side node to obtain a target model. A learning module 702 for performing federated learning with an FL client node based on the first request message and obtaining the target model. A sending module 703 for sending information about the target model to the model training function node.

[0197] Optionally, the first request message includes at least one of federated learning instruction information and a model identifier. The federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model. The model identifier is for uniquely identifying the target model.

[0198] Optionally, the first request message includes information about FL client nodes participating in federated learning.

[0199] Optionally, the model information sending device 700 further includes a determination module for determining FL client nodes participating in federated learning.

[0200] Optionally, the determination module is used for sending a node discovery request message for requesting FL client nodes participating in federated learning to a network repository function network element and receiving a response message sent by the network repository function network element, wherein the response message includes information about FL client nodes participating in federated learning.

[0201] Optionally, the information about the federated model including at least one piece of information among a model identifier, association indication information, a model file, and address information of the model file corresponding to the target model; The association learning instruction information is for instructing that the target model is a model obtained by association learning. The model identifier is for uniquely identifying the target model.

[0202] Optionally, the model identifier is obtained by the model training function node for the target model.

[0203] Optionally, the model information transmitting device 700 further includes an acquisition module for acquiring the model identifier for the target model.

[0204] The above model information transmitting device can improve the model training performance of the network node.

[0205] The model information transmitting device in the embodiments of the present application may be an electronic device, such as an electronic device equipped with an operating system, or a component in an electronic device, such as an integrated circuit or a chip. The electronic device may be a core network device, a network-side device, or a terminal, or other devices other than the terminal.

[0206] The model information transmitting device provided in the embodiments of the present application realizes each process realized in the method embodiment shown in FIG. 3 and can achieve the same technical effect. For the sake of brevity, detailed descriptions are omitted here.

[0207] Optionally, as shown in FIG. 8, an embodiment of the present application further provides a communication device 800. The communication device 800 includes a processor 801 and a memory 802. The memory 802 stores programs or commands executable by the processor 801. For example, when the communication device 800 is a first control network element, when the program or command is executed by the processor 801, each step of the embodiment of the above model information acquisition method or the above model information transmission method is realized, and the same technical effect can be achieved. Without repeating the description, the detailed description is omitted here.

[0208] An embodiment of the present application includes a processor and a communication interface. The processor or the communication interface is used to determine an FL server-side node. The communication interface is used to send a first request message for triggering the federated learning of the FL server-side node to obtain a target model to the FL server-side node, and to receive the information of the target model sent by the FL server-side node, and further provides a model training function node.

[0209] An embodiment of the present application includes a processor and a communication interface. The communication interface is used to receive a first request message sent by a model training function node. The first request message is for triggering the federated learning of the FL server-side node to obtain a target model. The processor or the communication interface is used to perform federated learning with an FL client node based on the first request message to obtain the target model. The communication interface is used to send the information of the target model to the model training function node, and further provides a server-side node.

[0210] Specifically, the embodiments of the present application further provide a network node. As shown in FIG. 9, the network node 900 includes a processor 901, a network interface 902, and a memory 903. Here, the network interface 902 is, for example, a common public radio interface (CPRI).

[0211] Specifically, the network node 900 of the embodiments of the present application further includes commands or programs stored in the memory 903 and executable by the processor 901. The processor 901 calls the commands or programs in the memory 903 to execute the methods executed by the respective modules shown in FIG. 6 or FIG. 7, and achieves the same technical effects. For the sake of avoiding repeated description, detailed description is omitted here.

[0212] In the embodiment where the above network node is a model training function node, the processor 901 or the network interface 902 is used to determine the federated learning FL server-side node, and the network interface 902 is used to send a first request message for triggering the federated learning of the FL server-side node to obtain a target model to the FL server-side node, and receive the information of the target model sent by the FL server-side node.

[0213] Optionally, the above-mentioned determining the FL server-side node includes sending a node discovery request message for requesting network nodes participating in the federated learning training to a network repository function network element, and receiving a response message sent by the network repository function network element, where the response message includes the information of the FL server-side node.

[0214] Optionally, the node discovery request message is including at least one of analysis identifier, region of interest AOI information, time of interest information, model description method information, model sharable information, model performance information, model algorithm information, model training speed information, federated learning instruction information, federated learning type information, FL server side node type instruction information, FL client node type instruction information, first service information, and second service information, the federated learning instruction information is for instructing that the network node requested by the request message should support federated learning; the first service information is for instructing that the network node requested by the request message should support the service of the federated learning server; the second service information is for instructing that the network node requested by the request message should support the service of the federated learning member.

[0215] Optionally, the federated learning type information indicates that the type of federated learning that the network node requested by the request message should support is horizontal federated learning and vertical federated learning, at least one of them.

[0216] Optionally, the response message includes information of N network nodes including the FL server side node, where N is a positive integer.

[0217] The information of each network node includes at least one of fully qualified domain name FQDN, identifier information, and address information.

[0218] Optionally, the N network nodes further include FL client nodes.

[0219] Optionally, the information of each network node is Further include type information for indicating the type of the network node, where the type is one of the FL server-side node and the FL client node.

[0220] Optionally, the first request message includes at least one of the federated learning instruction information and the model identifier, where the federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model, and the model identifier is for uniquely identifying the target model.

[0221] Optionally, the processor 901 or the network interface 902 further is used to obtain the model identifier.

[0222] Optionally, the first request message includes information about the FL client nodes participating in the federated learning.

[0223] Optionally, the information of the target model includes at least one of the model identifier, the federation instruction information, the model file, and the address information of the model file corresponding to the target model, where the federated learning instruction information is for indicating that the target model is a model obtained by federated learning, and the model identifier is for uniquely identifying the target model.

[0224] Optionally, the network interface 902 further is used by the model training function node to send the information of the target model to the model inference function node.

[0225] Optionally, before the model training function node determines the FL server-side node, the processor 901 further It is used to determine that the target model should be obtained by federated learning.

[0226] Optionally, the determination that the target model should be obtained by federated learning is when the model training function node determines that it cannot obtain all or part of the training data for generating the target model, including determining that the target model should be obtained by federated learning.

[0227] In an embodiment where the network node is an FL server-side node, the network interface 902 is to receive a first request message sent by a model training function node, where the first request message is for triggering the federated learning of the FL server-side node to obtain a target model, performing federated learning with an FL client node based on the first request message to obtain the target model, and sending information of the target model to the model training function node.

[0228] Optionally, the first request message includes at least one of federated learning instruction information and a model identifier, the federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model, and the model identifier is for uniquely identifying the target model.

[0229] Optionally, the first request message includes information of FL client nodes participating in federated learning.

[0230] Optionally, the processor 901 or the network interface 902 is further used to determine FL client nodes participating in federated learning.

[0231] Optionally, determining the FL client nodes that participate in federated learning includes: sending a node discovery request message for requesting FL client nodes that participate in federated learning to a network repository function network element; receiving a response message sent by the network repository function network element, where the response message includes information about FL client nodes that participate in federated learning.

[0232] Optionally, the information of the federated model includes: at least one piece of information among a model identifier, federation indication information, a model file, and address information of the model file, corresponding to the target model; the federated learning instruction information is for instructing that the target model is a model obtained by federated learning; the model identifier is for uniquely identifying the target model.

[0233] Optionally, the model identifier is obtained by the model training function node for the target model.

[0234] Optionally, the processor 901 is further used for: obtaining the model identifier for the target model.

[0235] It should be noted that in this embodiment, the model training function node and the FL server-side node are taken as examples to illustrate that they are core network elements.

[0236] Embodiments of the present application further provide a readable storage medium in which a program or command is stored, and when the program or command is executed by a processor, the steps of the model information acquisition method provided in the embodiments of the present application are realized, or the steps of the model information transmission method provided in the embodiments of the present application are realized.

[0237] The processor is the processor in the terminal described in the above embodiments. The readable storage medium includes computer-readable storage media such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk, or an optical disk.

[0238] The embodiments of the present application further provide a chip. The chip includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to execute a program or command to implement each process of the embodiments of the above model information acquisition method or model information transmission method, and the same technical effect can be achieved. For the sake of not repeating the description, the detailed description is omitted here.

[0239] It should be understood that the chip mentioned in the embodiments of the present application may also be referred to as a system-level chip, a system-on-chip, a chip system, or a system-on-chip, etc.

[0240] The embodiments of the present application further provide a computer program / program product. The computer program / program product is stored in a storage medium and is executed by at least one processor to implement each process of the embodiments of the above model information acquisition method or model information transmission method, and the same technical effect can be achieved. For the sake of not repeating the description, the detailed description is omitted here.

[0241] The embodiments of the present application further provide a model information transmission system including a model training function node and a server-side node. The model training function node can be used to execute the steps of the model information acquisition method provided in the embodiments of the present application, and the server-side node can be used to execute the steps of the model information transmission method provided in the embodiments of the present application.

[0242] It should be noted that in this specification, the term "comprising", "consisting of" or any other variations thereof is intended to include non-exclusive inclusion, such that a process, method, article or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such a process, method, article or apparatus. Unless otherwise specified, the elements limited by the phrase "comprising one..." do not exclude the further existence of the same other elements in the process, method, article or apparatus comprising the said element. It should also be pointed out that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order illustrated or considered, and may also include performing functions substantially simultaneously or in reverse order according to such functions. For example, the described method may be executed in an order different from that described, and various steps may be added, omitted, or combined. Also, the features described with reference to any example may be combined in other examples.

[0243] From the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above examples can be realized in the form of a combination of software and the necessary common hardware platform. Of course, it may also be realized by hardware, but in many cases the former is a more preferred embodiment. Based on such an understanding, the technical solution of this application, in essence or the part contributing to the prior art, can be implemented in the form of a computer software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a plurality of commands for causing a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of this application.

[0244] The embodiments of the present application have been described above with reference to the drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely exemplary and not restrictive. Based on the suggestions of the present application, many forms that those skilled in the art can achieve without departing from the spirit of the present application and the scope of protection of the claims all belong to the scope of protection of the present application.

Claims

1. The step in which the model training function node determines the federated learning FL server-side node; The step in which the model training function node transmits a first request message for triggering the federated learning of the FL server-side node to obtain a target model to the FL server-side node; The step in which the model training function node receives the information of the target model transmitted by the FL server-side node, and a model information acquisition method including the above steps.

2. The step in which the model training function node determines the FL server-side node is as follows: The step in which the model training function node transmits a node discovery request message for requesting network nodes participating in federated learning training to the network repository function network element; The step in which the model training function node receives a response message transmitted by the network repository function network element, and the response message includes information of the FL server-side node, and the method according to claim 1 including the above steps.

3. The node discovery request message: Includes at least one of an analysis identifier, area of interest AOI information, time of interest information, model description method information, model sharable information, model performance information, model algorithm information, model training speed information, federated learning instruction information, federated learning type information, FL server-side node type instruction information, FL client node type instruction information, first service information, and second service information; The federated learning instruction information is for instructing that the network node requested by the request message should support federated learning; The first service information is for instructing that the network node requested by the request message should support the service of the federated learning server; The second service information is for instructing that the network node requested by the request message should support the service of the federated learning member, and the method according to claim 2.

4. The federated learning type information is the type of federated learning that the network node requested by the request message should support: Horizontal federated learning; The method according to claim 3, which is for instructing that it is at least one of vertical federated learning.

5. The response message includes information of N network nodes including the FL server-side node, where N is a positive integer, The information of each network node is The method according to claim 2, including at least one of a fully qualified domain name FQDN, identifier information, and address information.

6. The method according to claim 5, wherein the N network nodes further include FL client nodes.

7. The information of each network node is The method according to claim 5 or claim 6, further including type information for instructing the type of the network node, and the type is one of an FL server-side node and an FL client node.

8. The first request message Includes at least one of federated learning instruction information and a model identifier, The federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model, The method according to any one of claims 1 to 6, wherein the model identifier is for uniquely identifying the target model.

9. The method according to claim 8, further including the step of the model training function node obtaining the model identifier.

10. The method according to any one of claims 1 to 6, wherein the first request message includes information of FL client nodes participating in federated learning.

11. The information of the target model is Includes at least one piece of information among a model identifier, federated instruction information, a model file, and address information of the model file corresponding to the target model, The federated learning instruction information is for instructing that the target model is a model obtained by federated learning, The method according to any one of claims 1 to 6, wherein the model identifier is for uniquely identifying the target model.

12. The method according to any one of claims 1 to 6, further including the step of the model training function node sending the information of the target model to the model inference function node.

13. Before the step of the model training function node determining the FL server-side node, The method according to any one of claims 1 to 6, further comprising the step of the model training function node determining that the target model should be obtained by federated learning.

14. The step in which the model training function node determines that the target model should be obtained by federated learning is as follows: The method according to claim 13, including the step of the model training function node determining that all or part of the training data for generating the target model cannot be obtained, and then determining that the target model should be obtained by federated learning.

15. A model information transmission method, including: a step in which a federated learning FL server-side node receives a first request message sent by a model training function node, where the first request message is for triggering the federated learning of the FL server-side node to obtain a target model; a step in which the FL server-side node performs federated learning with an FL client node based on the first request message and obtains the target model; a step in which the FL server-side node transmits information about the target model to the model training function node.

16. The first request message includes at least one of federated learning instruction information and a model identifier, the federated learning instruction information is for requesting the FL server-side node to trigger federated learning to obtain a target model, and the model identifier is for uniquely identifying the target model. The method according to claim 15.

17. The method according to claim 15 or claim 16, wherein the first request message includes information about FL client nodes participating in federated learning.

18. The method according to claim 15 or claim 16, further comprising the step of the FL server-side node determining FL client nodes participating in federated learning.

19. The step in which the FL server-side node determines FL client nodes participating in federated learning is as follows: The step in which the FL server-side node transmits a node discovery request message for requesting FL client nodes participating in federated learning to a network repository function network element. The step in which the FL server-side node receives a response message transmitted by the network repository function network element, wherein the response message includes information on FL client nodes participating in federated learning, and the method according to claim 18.

20. The information of the federated model is including at least one of information on a model identifier, federation instruction information, a model file, and address information of the model file corresponding to the target model, wherein the federated learning instruction information is for instructing that the target model is a model obtained by federated learning, wherein the model identifier is for uniquely identifying the target model, and the method according to claim 15 or claim 16.

21. The method according to claim 16, wherein the model identifier is obtained by the model training function node for the target model.

22. The method according to claim 20, further including the step in which the FL server-side node obtains the model identifier for the target model.

23. A first determination module for determining a federated learning FL server-side node; A first transmission module for transmitting a first request message for triggering the federated learning of the FL server-side node to obtain a target model to the FL server-side node; A receiving module for receiving information on the target model transmitted by the FL server-side node, and a model information acquisition device.

24. A receiving module for receiving a first request message transmitted by a model training function node, wherein the first request message is for triggering the federated learning of a federated learning FL server-side node to obtain a target model; A learning module for performing federated learning with an FL client node based on the first request message to obtain the target model; A transmission module for transmitting information on the target model to the model training function node, and a model information transmission device.

25. A model training function node including a processor and a memory, wherein a program or command executable by the processor is stored in the memory, and when the program or command is executed by the processor, the steps of the model information acquisition method according to any one of claims 1 to 14 are realized.

26. A server-side node including a processor and a memory, wherein a program or command executable by the processor is stored in the memory, and when the program or command is executed by the processor, the steps of the model information transmission method according to any one of claims 15 to 22 are realized.

27. A readable storage medium in which a program or command is stored, and when the program or command is executed by a processor, the steps of the model information acquisition method according to any one of claims 1 to 14 are realized, or the steps of the model information transmission method according to any one of claims 15 to 22 are realized.

Images