Online model fine tuning method and device based on edge intelligence and computer equipment

Abstract

The invention relates to an online model fine tuning method and device based on edge intelligence and computer equipment. The method comprises the following steps: receiving an online model fine tuning request uploaded by at least one terminal node to which an edge node belongs, and generating an online model fine tuning node group according to the at least one terminal node; according to the model fine tuning configuration file list, broadcasting model fine tuning feature signaling to each terminal node in the online model fine tuning node group; acquiring a subscription request signaling sent by any terminal node; and transmitting the subscribed model parameter configuration file to any terminal node, so that any terminal node performs online model fine tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model. By adopting the method, the model configuration efficiency in edge intelligence can be improved.

Description

Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to an edge intelligence-based online model fine-tuning method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] With the continuous improvement of the computing power of user devices and the increasing demand for massive data processing, network intelligence is expanding from cloud servers to the edge of the network, making edge intelligence a key driving factor for the application of the future sixth-generation mobile communication standard. Edge intelligence combines artificial intelligence models with edge computing technology. Relying on the edge computing architecture, data processing, model reasoning, and some training tasks are distributed on the edge nodes and terminal nodes of the network, rather than relying on traditional centralized cloud server computing. Edge intelligence reduces the delay in data transmission to the cloud by performing intelligent computing at edge nodes close to the data source, thereby improving response speed and privacy protection capabilities. Edge intelligence is widely used in smart homes, autonomous driving, industrial automation and other fields, and can realize real-time monitoring, anomaly detection, predictive maintenance and other functions.

[0003] When implementing edge intelligence in wireless networks, when there is a certain deviation between the deployment environment of the artificial intelligence model and its training environment, the data distribution learned by the artificial intelligence model in the training environment is inconsistent with the actual data distribution in the deployment environment, which will lead to a decrease in the performance of the artificial intelligence model. Therefore, it is necessary to configure the model based on the existing pre-trained model, that is, to adjust the model parameters using new data so that the model can adapt to new environments or tasks. However, with more and more user terminals accessing the edge of the network and people's increasing demand for artificial intelligence services, model configuration faces multiple challenges such as limited computing and communication resources and data security, resulting in low efficiency in model configuration in edge intelligence. Summary of the invention

[0004] Based on this, it is necessary to provide an edge intelligence-based online model fine-tuning method, device, computer equipment, computer-readable storage medium and computer program product that can improve the efficiency of model configuration in edge intelligence in response to the above-mentioned technical problems.

[0005] In a first aspect, the present application provides an online model fine-tuning method based on edge intelligence, which is applied to an edge node of a model fine-tuning system, wherein the model fine-tuning system also includes a terminal node and a network node; including:

[0006] Receiving an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, and generating an online model fine-tuning node group according to at least one of the terminal nodes;

[0007] broadcasting model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to the model fine-tuning configuration file list; the model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each of the terminal nodes;

[0008] Obtaining a subscription request signaling sent by any of the terminal nodes; the subscription request signaling is sent by any of the terminal nodes when a subscribed model parameter configuration file is determined based on the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling;

[0009] The subscribed model parameter configuration file is transmitted to any of the terminal nodes, so that any of the terminal nodes performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

[0010] In one embodiment, after obtaining the updated model, the method further includes:

[0011] Obtaining online model monitoring information sent by any of the terminal nodes; the online model monitoring information includes a model verification result of online verification of the updated model by any of the terminal nodes;

[0012] When the online model monitoring information indicates that the updated model meets the fine-tuning requirement condition, any terminal node is removed from the online model fine-tuning node group to obtain the updated online model fine-tuning node group.

[0013] In one embodiment, the model fine-tuning feature information includes at least one of the identification of the model parameter configuration file, computational complexity, storage complexity, implementation function, task type, expected performance gain, fine-tuning delay, and security level; after obtaining the updated online model fine-tuning node group, the method further includes:

[0014] According to the online model monitoring information, the subscribed model parameter configuration file is updated to obtain the updated model parameter configuration file, and the model fine-tuning feature information is updated to obtain the updated model fine-tuning feature information.

[0015] In one embodiment, after obtaining the updated model fine-tuning feature information, the method further includes:

[0016] Returning to the step of broadcasting model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to the model fine-tuning configuration file list, until a model fine-tuning termination condition is met;

[0017] The model fine-tuning termination condition includes that each of the terminal nodes is moved out of the online model fine-tuning node group or receives an online model fine-tuning termination signaling sent by the network node.

[0018] In one of the embodiments, before receiving the online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, the method further includes:

[0019] Generate model fine-tuning requirement information of the cell group according to model fine-tuning requirement information of each terminal node to which the edge node belongs;

[0020] Uploading the model fine-tuning requirement information of the cell group to the network node, so that the network node generates a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group;

[0021] Obtain the model fine-tuning configuration file list sent by the network node.

[0022] In a second aspect, the present application also provides an online model fine-tuning system based on edge intelligence, the system is used to implement the steps of the above method; the system includes a network node, an edge node and a terminal node;

[0023] The network node is used to generate a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group uploaded by the edge node; the model fine-tuning requirement information of the cell group is generated according to the model fine-tuning requirement information of each terminal node to which the edge node belongs;

[0024] The edge node is used to obtain the model fine-tuning configuration file list sent by the network node;

[0025] The terminal node is used to send an online model fine-tuning request to the edge node;

[0026] The edge node is configured to generate an online model fine-tuning node group according to at least one of the terminal nodes, and broadcast a model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to the model fine-tuning configuration file list;

[0027] The terminal node is used to determine the subscribed model parameter configuration file according to the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling, and send a subscription request signaling to the edge node;

[0028] The edge node is used to transmit the subscribed model parameter configuration file to the terminal node;

[0029] The terminal node is used to perform online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

[0030] In a third aspect, the present application further provides an online model fine-tuning device based on edge intelligence, which is applied to an edge node of a model fine-tuning system, wherein the model fine-tuning system further includes a terminal node and a network node; the device includes:

[0031] A receiving module, configured to receive an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, and generate an online model fine-tuning node group according to at least one of the terminal nodes;

[0032] A broadcast module, configured to broadcast model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to a model fine-tuning configuration file list; the model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each of the terminal nodes;

[0033] An acquisition module, used to acquire a subscription request signaling sent by any of the terminal nodes; the subscription request signaling is sent by any of the terminal nodes when a subscribed model parameter configuration file is determined based on the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling;

[0034] A transmission module is used to transmit the subscribed model parameter configuration file to any of the terminal nodes, so that any of the terminal nodes performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

[0035] In a fourth aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0036] In a fifth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, and the computer program implements the steps of the above method when executed by a processor.

[0037] In a sixth aspect, the present application also provides a computer program product, including a computer program, which implements the steps of the above method when executed by a processor.

[0038] The above-mentioned online model fine-tuning method, device, computer equipment, computer-readable storage medium and computer program product based on edge intelligence receive an online model fine-tuning request uploaded by at least one terminal node belonging to the edge node, and generate an online model fine-tuning node group according to at least one terminal node; broadcast model fine-tuning feature signaling to each terminal node in the online model fine-tuning node group according to the model fine-tuning configuration file list, wherein the model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each terminal node; obtain the subscription request signaling sent by any terminal node; wherein the subscription request signaling is sent by any terminal node when the subscribed model parameter configuration file is determined according to the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling; the subscribed model parameter configuration file is transmitted to any terminal node, so that any terminal node performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model. By receiving online model fine-tuning requests from terminal nodes, online model fine-tuning node groups are dynamically generated, and model fine-tuning feature signaling is broadcast according to the model fine-tuning configuration file list. Through the signaling interaction process between edge nodes and terminal nodes, it is ensured that the model configuration can be efficiently optimized according to actual needs. The terminal node can autonomously and flexibly select the adapted model parameter configuration file according to its own performance, environment and resource conditions, and perform online model fine-tuning on the pre-trained model, realizing flexible adaptation to time-varying environments and enhancing the generalization of artificial intelligence models on the edge side of the network. Therefore, customized artificial intelligence services can be provided to edge users based on general pre-trained models, while reducing data transmission volume and communication delays while improving model configuration efficiency and real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 An application environment diagram of an online model fine-tuning method based on edge intelligence in an embodiment;

[0041] Figure 2 A schematic diagram of a flow chart of an online model fine-tuning method based on edge intelligence in an embodiment;

[0042] Figure 3 A signaling interaction logic diagram of an online model fine-tuning method based on edge intelligence in an embodiment;

[0043] Figure 4 A schematic diagram of a flow chart of an online model fine-tuning method based on edge intelligence in another embodiment;

[0044] Figure 5 is a structural block diagram of an online model fine-tuning device based on edge intelligence in one embodiment;

[0045] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0047] The online model fine-tuning method based on edge intelligence provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown in the figure, the online model fine-tuning system based on edge intelligence includes a network node 100 , an edge node 102 and a terminal node 104 .

[0048] Among them, the network node 100 can manage and generate model parameter configuration files, maintain model versions, and configure signaling; the network node 100 also centrally stores various pre-trained models and model parameter configuration files for edge nodes 102 and terminal nodes 104 to retrieve; the network node 100 optimizes resource allocation and task scheduling by monitoring the operating status of the entire network. Exemplarily, the network node 100 has operations, administration, and maintenance (OAM), network data analytics function (NWDAF), third-party service platform function (Over-The-Top Server, OTT Server), network functions for network management, etc.

[0049] Among them, the edge node 102 is an intermediate layer connecting the user equipment (terminal node 104) and the core network (network node 100), which is used to share the computing tasks of the core network and provide low-latency services for the terminal nodes. The edge node 102 can be a node deployed close to the data source or user equipment, which can perform data processing, storage and some artificial intelligence computing tasks. For example, the edge node 102 may include a base station, an access network unit, an edge server, an industrial edge gateway or an edge computing device.

[0050] In one embodiment, the edge node 102 may include a centralized unit (CU) and a distributed unit (DU). Specifically, the centralized unit is an important component of the wireless access network, which is mainly responsible for higher-level protocol processing and resource management, and can be used for model transmission, signaling configuration, model update, etc. The centralized unit can be further divided into two parts: a control plane centralized unit (CU-CP) and a user plane centralized unit (CU-UP). Among them, the control plane centralized unit (CU-CP) can be responsible for the processing of control signaling, such as access management, switching management and connection maintenance of user equipment, and centralized management and coordination of multiple distributed units (DU). The user plane centralized unit (CU-UP) can be responsible for forwarding and processing user data, including the transmission of packet data, providing data plane processing, and assisting the distributed unit (DU) to complete efficient data transmission tasks. Specifically, the distributed unit is mainly responsible for low-level physical layer and real-time task processing, and performs specific tasks assigned by the centralized unit, which can be used for data collection, model transmission, etc.

[0051] Among them, the terminal node 104 is a user equipment (UE), and the terminal node 104 can be used for data collection, such as collecting environment and task-related data (such as sensor data, user behavior data, etc.). The terminal node 104 can also be used for model subscription and local model fine-tuning, that is, to adjust the local parameters of the pre-trained model according to the received model parameter configuration file to generate a model that adapts to the current environment or task; the terminal node 104 can also perform performance monitoring, monitor the operating status of the model in real time, including accuracy, delay and stability, etc., and feed back the monitoring information to the edge node 102.

[0052] It can be seen that the terminal node 104 may include a model fine-tuning function module, a model monitoring function module, a data acquisition function module, etc.; the edge node 102 may include a model fine-tuning function module, a model management function module, a model storage function module, etc.; the network node 100 may include a model management module, a model storage module, etc.

[0053] In a specific implementation, the network node 100 is used to generate a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group uploaded by the edge node 102; the model fine-tuning requirement information of the cell group is generated according to the model fine-tuning requirement information of each terminal node 104 to which the edge node 102 belongs;

[0054] The edge node 102 is used to obtain the model fine-tuning configuration file list sent by the network node 100;

[0055] The terminal node 104 is used to send an online model fine-tuning request to the edge node 102;

[0056] The edge node 102 is configured to generate an online model fine-tuning node group according to at least one terminal node 104, and broadcast a model fine-tuning feature signaling to each terminal node 104 in the online model fine-tuning node group according to the model fine-tuning configuration file list;

[0057] The terminal node 104 is used to determine the subscribed model parameter configuration file according to the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling, and send a subscription request signaling to the edge node 102;

[0058] The edge node 102 is used to transmit the subscribed model parameter configuration file to the terminal node 104;

[0059] The terminal node 104 is used to perform online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

[0060] In an exemplary embodiment, Figure 2 As shown in the figure, an online model fine-tuning method based on edge intelligence is provided. Figure 1 The edge node 102 in FIG. 1 is used as an example to illustrate, including:

[0061] Step S202: receiving an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, and generating an online model fine-tuning node group according to the at least one terminal node.

[0062] Among them, the online model fine-tuning request can be a request sent by the terminal node to the edge node during operation to express the terminal node's current demand for artificial intelligence model fine-tuning. The online model fine-tuning request may include the terminal node's current resource requirements (such as computing power, bandwidth, etc.), real-time requirements, and the type of artificial intelligence task that needs to be processed (such as natural language processing, image classification, target detection, etc.). In a specific implementation, the terminal node can report an online model fine-tuning request to the edge node at any time within each time period, and after the edge node responds to the request, participate in the online model fine-tuning process of the current time period or the next time period.

[0063] Optionally, the online model fine-tuning request can be sent in the form of downlink control information (Downlink Control Information, DCI), media access control layer control element (Media Access Control Control Element, MAC CE), and radio resource control signaling (Radio Resource Control Signaling, RRC). Among them, the downlink control information can be the signaling for transmitting control information between the user equipment and the base station, the media access control layer control element can be the control signaling for status reporting and command transmission between the user equipment and the base station, and the radio resource control signaling can be the signaling located in the radio resource control layer responsible for configuring and managing the radio link. Through these signaling, the online model fine-tuning request can be embedded in different layers of the protocol stack to ensure efficient transmission in different scenarios.

[0064] Optionally, the triggering method of the online model fine-tuning request includes at least one of the following: the terminal node determines whether online model fine-tuning needs to be performed based on model monitoring information, etc., and triggers an online model fine-tuning event, and sends an online model fine-tuning request to the edge node; the network node or edge node periodically sends model fine-tuning instructions to the terminal node, so that the terminal node generates an online model fine-tuning request.

[0065] In a specific implementation, the edge node may record the terminal nodes that send online model fine-tuning requests within the current time period, determine these terminal nodes as all terminal nodes that participate in online model fine-tuning within the current time period, and form an online model fine-tuning node group based on these terminal nodes.

[0066] In a specific implementation, the online model fine-tuning node group may include terminal nodes that have newly joined the online model fine-tuning process in the current time period, and terminal nodes that have joined the online model fine-tuning process in the previous time period and have not exited. The online model fine-tuning node group may record the information of each type of terminal nodes separately according to these two types of terminal nodes, or may record all terminal node information jointly without distinguishing between the two types.

[0067] Optionally, the online model fine-tuning node group records at least the terminal node identification (ID) of each terminal node, the cell / cell group to which it belongs, the timestamp of the terminal node joining the online model fine-tuning, whether it participates in joint fine-tuning, etc. Among them, the cell to which it belongs may refer to the area loaded by a base station in the wireless network, and the cell group may refer to a set of multiple adjacent or functionally related cells; whether to participate in joint fine-tuning may refer to that multiple terminal nodes can collaboratively complete the model fine-tuning task and can share model update information or fine-tuning results.

[0068] In a specific implementation, the online model fine-tuning peer group supports terminal nodes to join and leave the group at any time, and updates the information of the online model fine-tuning peer group after each terminal node joins and leaves the group.

[0069] Step S204: broadcast model fine-tuning feature signaling to each terminal node in the online model fine-tuning node group according to the model fine-tuning configuration file list.

[0070] In a specific implementation, the edge node can filter out the model fine-tuning feature information from the model fine-tuning configuration file list according to the model fine-tuning configuration file list, generate the model fine-tuning feature signaling based on the model fine-tuning feature information, and broadcast the model fine-tuning feature signaling to each terminal node in the online model fine-tuning node group, and the model fine-tuning feature signaling carries the model fine-tuning feature information. Among them, the model fine-tuning configuration file list may include multiple model parameter configuration files and the corresponding model fine-tuning feature information. In a specific implementation, the model fine-tuning configuration file list can be generated by the network node based on the model fine-tuning requirement information of the terminal node and sent to the edge node on the basis of the pre-trained model.

[0071] The model parameter configuration file may include parameter configuration information of the pre-trained model, and the parameter configuration information may be used to guide online model fine-tuning of the pre-trained model. Each model parameter configuration file may correspond to unique model fine-tuning feature information.

[0072] The model fine-tuning feature information may be used to describe feature information related to executing model fine-tuning in a model parameter configuration file.

[0073] The model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each terminal node. Optionally, the model fine-tuning requirement information may include computing power configuration information related to the corresponding terminal node that can be used for model fine-tuning, storage space capacity information that can be used to store the model, service quality performance requirements for model fine-tuning, latency tolerance, application scenarios, data types or data formats, and other historical data related to model fine-tuning.

[0074] In one of the embodiments, the model fine-tuning feature information can be stored in the model fine-tuning configuration file list of the edge node in the manner of a hash table. The file identification ID of each model parameter configuration file is used as a keyword (key), and a mapping function is designed to map the keyword to the storage address (value) of the corresponding model fine-tuning feature information, thereby determining the unique storage address of the model fine-tuning feature information according to each file identification ID, and obtaining a one-to-one correspondence between multiple key-value pairs to ensure efficient storage and retrieval.

[0075] In one of the embodiments, the model fine-tuning feature information includes at least one of the identification of the model parameter configuration file, computational complexity, storage complexity, implementation function, task type, expected performance gain, fine-tuning delay, and security level.

[0076] Among them, the identifier of the model parameter configuration file can be the unique ID of the model parameter configuration file, which is used to distinguish each model parameter configuration file; the computational complexity indicates the size of the computing resources required by the model parameter configuration file, including the number of CPU / GPU processors, number of cycles, amount of floating-point operations, etc. required to perform model fine-tuning; the storage complexity indicates the size of the storage resources required by the model parameter configuration file, including the amount of disk space occupied, storage format, whether compression processing is supported, etc.; the implementation function characterizes the input and output types of the model parameter configuration file, such as the data type of the input data, the data format of the output data, whether multimodal data is input, etc.; the task type indicates the type of task that can be executed by the model parameter configuration file, including natural language processing, image classification, and sentiment analysis. classification, time series prediction, target detection, etc.; the expected performance gain represents the expected performance gain of the model parameter profile in terms of quality of service (QoS), energy efficiency, system performance, etc., and is related to the self-assessment results of the network nodes; the fine-tuning delay represents the processing time of the model parameter profile when performing the fine-tuning task, which is generally related to the type of task performed, such as the time required for the model parameter profile to process a data sample, the time required to process a batch of data samples, etc.; the security level represents the confidentiality level of the model parameter profile. If its model structure or design method has significant autonomy and innovation and has a high confidentiality level, in order to protect the security of data privacy, it will be restricted from being distributed to terminal nodes or only distributed to terminal nodes that comply with security regulations.

[0077] Exemplarily, model fine-tuning feature information may be as shown in Table 1.

[0078] Table 1

[0079]

[0080] In the specific implementation, the edge node can determine the corresponding model fine-tuning application scenario according to the online model fine-tuning request sent by the terminal node. For different model fine-tuning application scenarios, the model fine-tuning feature information screened out by the edge node is different, so the generated model fine-tuning feature signaling is different. The model fine-tuning application scenario can be the specific use occasion or task requirement of the artificial intelligence model in different practical applications. Since different scenarios have different requirements for the functions, performance, resource consumption, etc. of the model, the edge node needs to screen out the most suitable model fine-tuning feature information according to these scenario characteristics. For example, the terminal node may need different types of model fine-tuning algorithms for different application tasks, such as speech recognition, user behavior prediction, target detection, path planning, equipment anomaly detection, and predictive maintenance, which all correspond to different types of model fine-tuning algorithms; for another example, the computing power, storage space, bandwidth and other resources of the terminal node will limit the complexity and fine-tuning method of the model, such as high-performance terminal nodes can run complex fine-tuning algorithms, and low-performance terminal nodes can run lightweight models. Therefore, for these different model fine-tuning application scenarios, the edge node can screen out different model fine-tuning feature information, thereby broadcasting different model fine-tuning feature signaling.

[0081] In one embodiment, the edge node generates a model fine-tuning feature signaling based on the model fine-tuning feature information, and the model fine-tuning feature signaling may include the model fine-tuning feature information and may also implicitly or explicitly include a model fine-tuning configuration file list. Implicitly means that the model fine-tuning feature information already includes the identification ID of the model parameter configuration file, and the edge node does not need to transmit it separately; explicitly means that the edge node transmits the model fine-tuning configuration file list to the terminal node separately once.

[0082] In a specific implementation, the edge node broadcasts the model fine-tuning feature signaling to each terminal node in the online model fine-tuning node group. Optionally, the model fine-tuning feature signaling can be transmitted by the control plane centralized unit in the edge node through the downlink control information (DCI) or the system information broadcast (SIB), or it can be sent by the user plane centralized unit in the edge node to the user plane centralized unit through the data radio bearer (DRB). The model fine-tuning feature signaling can be transmitted after being attached to the downlink control information (DCI), or it can be transmitted within the downlink control information (DCI) field.

[0083] Step S206: Obtain a subscription request signaling sent by any terminal node.

[0084] The subscription request signaling is sent by any terminal node when the subscribed model parameter configuration file is determined based on the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling. The subscription request signaling is a signaling sent by the terminal node to the edge node to request a specific model parameter configuration file.

[0085] The node operation information includes at least one of performance information, environment information and resource information. The terminal node can determine the model parameter configuration file to be subscribed according to the matching of the node operation information of the current time period and the model fine-tuning feature information.

[0086] The model parameter configuration file includes parameter configuration information of the pre-trained model. The model parameter configuration file may include configurations optimized for specific tasks or terminal node requirements. The model parameter configuration file may define parameter adjustment ranges, optimization targets, training strategies, etc.

[0087] In one embodiment, the model parameter configuration file includes improvements in the structure, function, number of model parameters, connection method, etc. of the pre-trained model, and can change some modules of the original backbone model or add new modules based on it.

[0088] In one embodiment, the pre-trained model can be a large language model or a smaller general neural network model, such as a convolutional neural network CNN, a recurrent neural network RNN, a fully connected network MLP, a generative adversarial network GAN, etc.

[0089] In a specific implementation, the terminal node selects one or more model parameter configuration files based on the performance information, environment information, and resource information in the current time period. Optionally, the granularity of the time period can be a time slot, a half frame, a frame, or a longer time interval, which is uniformly configured by the edge node and notified to the terminal node.

[0090] Optionally, performance information may refer to performance requirements itself, which include requirements for performance indicators such as accuracy, convergence speed, energy consumption, latency, throughput, spectrum efficiency, and quality of service (QoS) related to model fine-tuning; environmental information may refer to environmental conditions, which include wireless channel conditions, data arrival conditions, business processing conditions, etc.; resource information may refer to available resources, which may include computing resources, communication resources, storage resources, etc. available to local devices.

[0091] In one embodiment, the method for a terminal node to select a model parameter configuration file to subscribe to may include: the terminal node determines the application scenario, performance requirement threshold, etc. in which it needs to perform model fine-tuning based on the performance information, environmental information, and resource information in the current time period; based on the model fine-tuning configuration file list, the terminal node selects the model fine-tuning feature information corresponding to each model parameter configuration file that meets its own application scenario, compares it with its own performance requirement threshold, obtains the score of each model parameter configuration file, and the model parameter configuration file that does not meet its own application scenario is not scored; selects the N configuration files with the highest scores from all model parameter configuration files as the model parameter configuration files to be subscribed by this terminal node, and forms a model parameter configuration file subscription list. Optionally, the value of N is configured by the terminal node itself, and can also be configured by the edge node through downlink control information or the media access control layer control unit.

[0092] In one of the embodiments, the model parameter configuration file subscription list may correspond to the model fine-tuning configuration file list issued by the edge node, and be arranged in the same order to maintain alignment at the edge node and the terminal node. A 1-bit flag may be added after each file identification ID in the model fine-tuning configuration file list to obtain the model parameter configuration file subscription list. In the model parameter configuration file subscription list, the flag position of the subscribed model parameter configuration file may be set to 1, and the flag positions of the remaining modules may be set to 0.

[0093] In one of the embodiments, the subscription request signaling may include a subscription list of model parameter configuration files related to the terminal node, and may also include information such as timestamp, subscription duration, subscription method, supported storage formats, etc.

[0094] Optionally, the subscription request signaling may be carried via control plane uplink control information (UCI) signaling or user plane data radio bearer (DRB) bearer.

[0095] Step S208: The subscribed model parameter configuration file is transmitted to any terminal node, so that any terminal node performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

[0096] In a specific implementation, the edge node selects the model parameter configuration file it subscribes to and transmits it to the corresponding terminal node based on the subscription request signaling reported by the terminal node. The model parameter configuration file may include parameter configuration information of the pre-trained model, which can be used to guide how to fine-tune and configure the pre-trained model.

[0097] Optionally, the edge node may transmit a subscribed model parameter profile based on a radio bearer (DRB) and indicate the allocated time-frequency domain transmission resources using downlink control information (DCI).

[0098] In a specific implementation, the method for transmitting a subscribed model parameter configuration file may include: the edge node transmits the model file of the model parameter configuration file from the centralized unit CU to which the terminal node belongs to the distributed unit DU through the F1 interface, and transmits it to the terminal node through the Uu interface, wherein the F1 interface is a communication interface connecting the distributed unit DU and the centralized unit CU, and the Uu interface is a wireless interface between the terminal device and the base station; or, the edge node transmits the model file of the model parameter configuration file from the centralized unit CU to which other non-terminal nodes belong to the centralized unit CU to which the terminal node currently belongs through the Xn interface, and then transmits the model file from the centralized unit CU to the distributed unit DU through the F1 interface, and transmits it to the terminal node through the Uu interface, wherein Xn is an interface for connecting different base stations; or, the edge node transmits the model file of the model parameter configuration file from the distributed unit DU to the terminal node through the Uu interface.

[0099] In one embodiment, the edge node may transmit the customized fine-tuning mode based on the storage format of the model parameter configuration file. The storage format of the model parameter configuration file may be an extensible format such as .h5, .pb, .onnx, or a serialized storage format such as .pt, .json, .pkl. Different terminal nodes may support different deep learning frameworks and thus support different storage formats. Therefore, the edge node may save different storage formats of the same model parameter configuration file and send model files of customized fine-tuning modes in different storage formats based on different terminal nodes.

[0100] In one of the embodiments, the edge node may participate in the decision-making process of module subscription, for example, based on factors such as the global quality of service (QoS) requirements of the cell / cell group, global energy consumption requirements, computing power scheduling capabilities, model parameter profile utilization, stability requirements, generalization requirements, etc., to adjust the module subscription results of the terminal node and ensure that the initial module subscription request of the terminal node is met to the greatest extent.

[0101] In the specific implementation, the terminal node uses the online fine-tuning dataset to perform model fine-tuning based on the subscribed model parameter configuration file transmitted by the edge node to obtain the updated model. The model fine-tuning algorithm can use the gradient descent algorithm to determine the update direction of the model by the gradient descent direction of the model parameters in the model parameter configuration file on each data sample or a batch of data samples, and update the model parameters of the pre-trained model by setting the update step size.

[0102] In the above-mentioned online model fine-tuning method based on edge intelligence, an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs is received, and an online model fine-tuning node group is generated according to at least one terminal node; according to the model fine-tuning configuration file list, a model fine-tuning feature signaling is broadcast to each terminal node in the online model fine-tuning node group, wherein the model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each terminal node; a subscription request signaling sent by any terminal node is obtained; wherein the subscription request signaling is sent by any terminal node when the subscribed model parameter configuration file is determined according to the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling; the subscribed model parameter configuration file is transmitted to any terminal node, so that any terminal node performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model. By receiving online model fine-tuning requests from terminal nodes, online model fine-tuning node groups are dynamically generated, and model fine-tuning feature signaling is broadcast according to the model fine-tuning configuration file list. Through the signaling interaction process between edge nodes and terminal nodes, it is ensured that the model configuration can be efficiently optimized according to actual needs. The terminal node can autonomously and flexibly select the adapted model parameter configuration file according to its own performance, environment and resource conditions, and perform online model fine-tuning on the pre-trained model, realizing flexible adaptation to time-varying environments and enhancing the generalization of artificial intelligence models on the edge side of the network. Therefore, customized artificial intelligence services can be provided to edge users based on general pre-trained models, while reducing data transmission volume and communication delays while improving model configuration efficiency and real-time performance.

[0103] In another embodiment, after obtaining the updated model, it also includes: obtaining online model monitoring information sent by any terminal node; the online model monitoring information includes the model verification result of any terminal node performing online verification on the updated model; when the online model monitoring information indicates that the updated model meets the fine-tuning requirements, any terminal node is removed from the online model fine-tuning node group to obtain an updated online model fine-tuning node group.

[0104] In the specific implementation, the terminal node collects online data in real time and divides the collected data into an online fine-tuning data set and an online verification data set. The division ratio is determined by the terminal node itself or configured by the edge node. Based on the subscribed model parameter configuration file transmitted by the edge node, the terminal node uses the online fine-tuning data set to perform model fine-tuning to obtain an updated model. The model fine-tuning algorithm can use a gradient descent algorithm to determine the update direction of the model by the gradient value descent direction of the model parameters in the model parameter configuration file on each data sample or a batch of data samples, and update the model parameters of the pre-trained model by setting the update step size. The terminal node performs online verification on the updated model based on the model parameters of the updated model and the verification results obtained for each data sample in the online verification data set, and compares them with the label value to obtain the online verification performance on the data sample, and finally calculates the average value on all online verification data samples to obtain the online verification performance based on the online verification data set, and obtains the model verification result.

[0105] In one of the embodiments, if a terminal node subscribes to multiple model parameter configuration files, multiple updated model parameter configuration files are used to obtain the online verification performance based on the verification results on the online verification data set, and the model verification results are obtained by averaging them or using a specific model integration algorithm. The model integration algorithm is determined autonomously by the terminal node.

[0106] In the specific implementation, the terminal node executes the online model monitoring mechanism, monitors the online verification performance in real time, and obtains the model verification results. The model verification results may include at least one of the loss value, gradient value, convergence speed, accuracy, latency, throughput, and spectrum efficiency of the online model verification, and may also include model usage feedback, model monitoring indicator description, online model fine-tuning log and other information. The terminal node can generate online model monitoring information based on these model verification results and upload it to the edge node.

[0107] Among them, the online model monitoring information may include the above-mentioned model verification results, that is, the feedback information generated after the terminal node verifies the updated model, including model performance, adaptability and task completion status, etc., which can be used to evaluate the effectiveness of the updated model and model parameter configuration file.

[0108] Optionally, the terminal node can report the online model monitoring information to the edge node via uplink control information (UCI) signaling or user plane data radio bearer (DRB). The reporting period can be set by the edge node or the terminal node, but it must be ensured that the reporting is completed within this time period.

[0109] In one of the embodiments, the online model monitoring information corresponds to the model parameter configuration file. If the terminal node subscribes to multiple model parameter configuration files, it reports the online model monitoring information corresponding to the multiple modules respectively. For example, a list is formed in the order of the identification ID of the model parameter configuration file, that is, [ID of module 1: online model monitoring information of module 1,..., ID of module M: ​​online model monitoring information of module M].

[0110] In the specific implementation, since the online model monitoring information can reflect the model fine-tuning process and model configuration results, if the online model monitoring information represents that the updated model meets the fine-tuning requirements, it can be said that the model configuration has been successful and meets the requirements of the fine-tuning task, thereby triggering the online model fine-tuning exit process of the terminal node.

[0111] Optionally, the fine-tuning requirement conditions may include at least one of the following conditions: the online model verification loss value or convergence speed is less than a preset threshold; the online model verification accuracy is greater than a preset threshold, or the change in T consecutive rounds is less than a preset threshold; the online model verification delay, throughput or spectrum efficiency is less than a preset threshold. The above preset thresholds and T values ​​can be configured by the edge node system or the terminal node itself.

[0112] In one embodiment, the terminal node triggers the online model fine-tuning exit process, sends an online model fine-tuning exit request to the edge node, and exits the online model fine-tuning node group. The edge node responds to the online model fine-tuning exit request, no longer sends the model parameter configuration file model fine-tuning feature information to the terminal node in the next time period, and updates the online model fine-tuning node group.

[0113] In one embodiment, the online model fine-tuning exit process of any terminal node can be triggered by the edge node according to the online model monitoring information reported by the terminal node, and the online model fine-tuning exit signaling is sent to the corresponding terminal node. When the terminal node feedbacks the confirmation information, the edge node removes it from the online model fine-tuning node group, no longer sends the model parameter configuration file model fine-tuning feature information to the terminal node in the next time period, and updates the online model fine-tuning node group.

[0114] The technical solution of this embodiment, through the introduction of online model monitoring information, can verify in real time whether the updated model in the terminal node meets the fine-tuning requirements, realize dynamic performance evaluation during the fine-tuning process, and enhance adaptability to complex scenarios and task requirements; when a terminal node verifies that its updated model meets the fine-tuning requirements, it is removed from the online model fine-tuning node group, which can reduce the resource allocation to the node that has completed the fine-tuning, allocate more resources to other nodes that need fine-tuning, and dynamically remove the nodes that have completed the task to avoid redundancy in system resource allocation; after the terminal node completes fine-tuning and meets the requirements, it is promptly removed from the node group, which can speed up the iteration efficiency of the fine-tuning task. The dynamic addition and removal mechanism of the terminal node adapts to the characteristics of dynamic changes in the status of network nodes and terminal nodes in the edge intelligent system, and enhances the real-time response capability.

[0115] In another embodiment, after the updated online model fine-tuning point group, it also includes: updating the subscribed model parameter configuration file according to the online model monitoring information to obtain the updated model parameter configuration file, and updating the model fine-tuning feature information to obtain the updated model fine-tuning feature information.

[0116] The model fine-tuning feature information includes at least one of the identification of the model parameter configuration file, computational complexity, storage complexity, implementation function, task type, expected performance gain, fine-tuning delay, and security level.

[0117] The edge node can improve the existing model parameter configuration file based on the model deficiencies fed back in the monitoring information. For example, updating the subscribed model parameter configuration file may include adjusting the model parameter range to make it more suitable for specific scenarios or task requirements.

[0118] In a specific implementation, the edge node can weight the sum of the online model verification gradient values ​​reported by the terminal nodes in all online model fine-tuning node groups according to the gradient value information in the online model monitoring information, and use the gradient descent method to update the model parameters contained in the subscribed model parameter configuration file. Alternatively, the edge node can increase or decrease the number of model parameters contained in the subscribed model parameter configuration file according to a certain ratio based on the online model verification delay information reported by the terminal nodes in all online model fine-tuning node groups for the subscribed model parameter configuration file, thereby reducing the transmission delay of the online model verification.

[0119] Accordingly, if the model parameter configuration file is updated, the corresponding model fine-tuning feature information is also updated.

[0120] The technical solution of this embodiment dynamically updates the model parameter configuration file and its feature information by acquiring online monitoring information in real time, and can quickly optimize the module performance according to the latest feedback from the terminal node, so that the model always adapts to the actual application scenario; by updating the subscribed fine-tuning module, unnecessary repeated debugging and performance degradation are avoided, ensuring that the optimization process of the model is more accurate and efficient; by updating the feature information, the edge node can more accurately assign the fine-tuning module to the terminal node that meets the requirements, so that the edge intelligent system can quickly respond to changes in terminal node requirements and adapt to complex scenarios of multiple tasks and multiple environments.

[0121] In another embodiment, after obtaining the updated model fine-tuning feature information, it also includes: returning to the step of broadcasting model fine-tuning feature signaling to each terminal node in the online model fine-tuning node group according to the model fine-tuning configuration file list until the model fine-tuning termination condition is met; wherein the model fine-tuning termination condition includes that each terminal node is moved out of the online model fine-tuning node group or receives an online model fine-tuning termination signaling issued by the network node.

[0122] Optionally, the online model fine-tuning termination signaling sent by the network node can be forwarded to the edge node through a dedicated control plane interface. The edge node can send the online model fine-tuning termination signaling to all terminal nodes, and can transmit the online model fine-tuning termination signaling by means of downlink control information, media access control layer control elements, and radio resource control signaling.

[0123] In the specific implementation, the iteration of the fine-tuning process will continue until the model fine-tuning termination condition is met. When the terminal node confirms that its model meets the fine-tuning requirements through online model verification, it will be removed from the fine-tuning node group one by one. Once the online model fine-tuning node group is empty, it means that the fine-tuning tasks of all terminal nodes have been completed, and the process is terminated at this time; or, the network node can actively send a termination signal according to global requirements (such as system resource reallocation, task priority adjustment, etc.) to force the end of the fine-tuning process.

[0124] In the iterative process of fine-tuning, after obtaining the updated model fine-tuning configuration file list, return to the step of broadcasting the model fine-tuning feature signaling to each terminal node in the online model fine-tuning node group according to the model fine-tuning configuration file list, and dynamically update the model fine-tuning feature signaling to ensure that the terminal node can obtain the model parameter configuration file that best suits the current state in each iteration. Through multiple iterations, the model performance is gradually optimized to finally meet the task requirements.

[0125] The technical solution of this embodiment optimizes the model fine-tuning process through dynamic iteration, ensuring that the terminal node can continuously receive the most suitable model parameter configuration file and gradually meet the task requirements; when the fine-tuning termination conditions are met, the process ends naturally, and the model fine-tuning task can be efficiently completed in a dynamic environment and multi-task scenario, while optimizing resource allocation.

[0126] In another embodiment, before receiving an online model fine-tuning request uploaded by at least one terminal node, it also includes: generating model fine-tuning requirement information of the cell group based on the model fine-tuning requirement information of each terminal node to which the edge node belongs; uploading the model fine-tuning requirement information of the cell group to the network node, so that the network node generates a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group; and obtaining the model fine-tuning configuration file list sent by the network node.

[0127] Among them, the model fine-tuning requirement information may include the computing power configuration information related to the corresponding terminal node that can be used for model fine-tuning, storage space capacity information that can be used to store the model, service quality performance requirements for model fine-tuning, latency tolerance, application scenarios, data types or data formats, and other historical data related to model fine-tuning.

[0128] The model fine-tuning requirement information of the cell group may include the model fine-tuning requirement information of the cell and the model fine-tuning requirement information of the cell group. The cell may refer to an area in a wireless network loaded by a base station, and the cell group may refer to a collection of multiple adjacent or functionally related cells. In addition to the model fine-tuning requirement information of each terminal node, the model fine-tuning requirement information of the cell group also includes the quantity requirements, stability requirements, generalization requirements, and other information of the models deployed in the cell / cell group.

[0129] In one embodiment, the model parameter configuration file includes improvements in the structure, function, number of model parameters, connection mode, etc. of the pre-trained model, which can change some modules of the original backbone model or add new modules on its basis. Optionally, the model parameter configuration file can be implemented using technologies such as low-rank adaptation (LoRA), adapter, prefix tuning, prompt tuning, etc.

[0130] In one of the embodiments, the pre-trained model can be a large language model, such as a Transformer model, a generative pre-trained transformer model, a decoding-enhanced BERT model, etc., or it can be a smaller-scale general neural network model, such as a convolutional neural network CNN, a recurrent neural network RNN, a fully connected network MLP, a generative adversarial network GAN, etc.

[0131] In a specific implementation, the network node can generate a corresponding model parameter configuration file based on the pre-trained model according to the model fine-tuning requirement information of the cell group, generate corresponding model fine-tuning feature information according to the feature information related to model fine-tuning in the model parameter configuration file, and generate a model fine-tuning configuration file list based on the model parameter configuration file and the model fine-tuning feature information. Exemplarily, the network node generates a corresponding model parameter configuration file based on the pre-trained model by generating a model parameter configuration file through low-rank adaptation (LoRA) technology. Specifically, the weight matrix of the pre-trained model can be low-rank decomposed into two matrices with fewer parameters, and the size of the parameters of the model parameter configuration file can be controlled by adjusting the rank of the matrix to form a series of model parameter configuration files for matching the performance requirements of different terminal nodes.

[0132] Optionally, the network node may transmit the model parameter configuration file to the edge node via a user plane data radio bearer (DRB), and transmit corresponding configuration information via dedicated control signaling.

[0133] In one of the embodiments, the edge node sorts the model parameter configuration files based on the model parameter configuration files transmitted by the network nodes according to at least one of the module parameter size, model structure complexity, the customized design method used, stability, generalization, etc., records the identification ID information of each module in sequence, and generates a model fine-tuning configuration file list.

[0134] The technical solution of this embodiment is that before the terminal node sends a fine-tuning request, the edge node and the network node collaborate to generate a fine-tuning module, which reduces the response delay; aggregates the demand information of the zone group to avoid the terminal node from transmitting the demand individually, saving communication resources; the edge node is only responsible for distribution, and the network node concentrates on completing complex model optimization tasks, which improves the overall utilization efficiency of system resources; the model parameter configuration file is generated according to the zone group demand, and can adapt to multiple terminal nodes at the same time, reducing repeated calculations; by dynamically maintaining the model fine-tuning configuration file list, it supports the different demand changes of terminal nodes and adapts to complex multi-task scenarios. In summary, through the collaboration of edge nodes and network nodes, a model fine-tuning configuration file list that adapts to the zone group demand is generated in advance, so that subsequent online model fine-tuning requests can be responded to quickly, which not only optimizes resource allocation, but also improves the efficiency and adaptability of model fine-tuning, and is suitable for processing complex scenarios with multiple terminals and multiple tasks.

[0135] In order to facilitate the understanding of those skilled in the art, Figure 3 A signaling interaction logic diagram of an online model fine-tuning method based on edge intelligence is provided by way of example.

[0136] The edge intelligence-based online model fine-tuning method of the embodiment of the present application can be oriented to the edge intelligence of the fifth-generation mobile communication standard, the sixth-generation mobile communication standard and other communication standards, and provide customized artificial intelligence services for edge users based on the general pre-trained model. In each time period, the edge device independently decides to deploy and execute the model fine-tuning algorithm according to its own performance requirements, environmental conditions and available resources in the current time period, thereby achieving flexible adaptation to the time-varying environment and enhancing the generalization of the artificial intelligence model on the edge side of the network. By proposing a cloud-edge-end collaborative model fine-tuning configuration method and a signaling interaction process, a model parameter configuration file generation and deployment strategy is proposed, and a signaling for characterizing model fine-tuning feature information is designed, which is suitable for a variety of typical wireless artificial intelligence use case scenarios. This method provides a solution for edge devices to perform online model fine-tuning, provides low-latency, customized model fine-tuning services for downstream edge devices based on cloud pre-trained models, and effectively addresses the problem of data distribution deviation caused by time-varying environments in online scenarios. By designing an online model fine-tuning configuration method and a signaling interaction mechanism, customized model fine-tuning services are provided for edge devices. A strategy for generating and deploying a model parameter configuration file is designed, a signaling that complies with the Third Generation Partnership Project (3GPP) standard for characterizing model fine-tuning feature information is defined, and the corresponding implementation process is given. An online model fine-tuning implementation method is designed, in which the terminal node independently decides to deploy and execute the model fine-tuning algorithm in each time period, thereby improving the adaptability of edge devices to time-varying environments while effectively protecting user privacy.

[0137] In another embodiment, Figure 4 As shown in the figure, an online model fine-tuning method based on edge intelligence is provided. Figure 1 Taking the edge node 102 in the example as an example, the following steps are included:

[0138] S402 , generating model fine-tuning requirement information of a cell group according to model fine-tuning requirement information of each terminal node to which the edge node belongs.

[0139] S404, uploading the model fine-tuning requirement information of the cell group to the network node, so that the network node generates a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group.

[0140] S406, obtaining a model fine-tuning configuration file list sent by the network node.

[0141] S408, receiving an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, and generating an online model fine-tuning node group according to the at least one terminal node.

[0142] S410 , broadcasting model fine-tuning feature signaling to each terminal node in the online model fine-tuning node group according to the model fine-tuning configuration file list.

[0143] The model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each terminal node.

[0144] S412: Obtain a subscription request signaling sent by any terminal node.

[0145] The subscription request signaling is sent by any terminal node when the subscribed model parameter configuration file is determined based on the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling.

[0146] S414, transmitting the subscribed model parameter configuration file to any terminal node, so that any terminal node performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

[0147] S416, obtaining online model monitoring information sent by any terminal node.

[0148] The online model monitoring information includes the model verification result of online verification of the updated model by any terminal node.

[0149] S418, when the online model monitoring information indicates that the updated model meets the fine-tuning requirement conditions, any terminal node is removed from the online model fine-tuning node group to obtain an updated online model fine-tuning node group.

[0150] S420, updating the subscribed model parameter configuration file according to the online model monitoring information to obtain an updated model parameter configuration file, and updating the model fine-tuning feature information to obtain an updated model fine-tuning feature information.

[0151] The model fine-tuning feature information includes at least one of the identification of the model parameter configuration file, computational complexity, storage complexity, implementation function, task type, expected performance gain, fine-tuning delay, and security level.

[0152] S422, determine whether the model fine-tuning termination condition is met, if so, end the process; if not, return to step S410.

[0153] The model fine-tuning termination condition includes that each terminal node is moved out of the online model fine-tuning node group or receives an online model fine-tuning termination signaling sent by a network node.

[0154] It should be noted that the specific limitations of the above steps can be found in the specific limitations of an online model fine-tuning method based on edge intelligence mentioned above.

[0155] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0156] Based on the same inventive concept, the embodiment of the present application also provides an edge intelligence-based online model fine-tuning device for implementing the edge intelligence-based online model fine-tuning method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more edge intelligence-based online model fine-tuning device embodiments provided below can refer to the limitations of the edge intelligence-based online model fine-tuning method above, and will not be repeated here.

[0157] In an exemplary embodiment, Figure 5 As shown, an online model fine-tuning device based on edge intelligence is provided, which is applied to the edge node of the model fine-tuning system. The model fine-tuning system also includes terminal nodes and network nodes; including:

[0158] The receiving module 510 is used to receive an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, and generate an online model fine-tuning node group according to at least one of the terminal nodes.

[0159] A broadcast module 520 is used to broadcast model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to a model fine-tuning configuration file list; the model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each of the terminal nodes;

[0160] The acquisition module 530 is used to obtain the subscription request signaling sent by any of the terminal nodes; the subscription request signaling is sent by any of the terminal nodes when the subscribed model parameter configuration file is determined based on the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling.

[0161] The transmission module 540 is used to transmit the subscribed model parameter configuration file to any of the terminal nodes, so that any of the terminal nodes can perform online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

[0162] In one of the embodiments, the edge intelligence-based online model fine-tuning device also includes a removal module, which is specifically used to obtain online model monitoring information sent by any of the terminal nodes; the online model monitoring information includes a model verification result of online verification of the updated model by any of the terminal nodes; when the online model monitoring information indicates that the updated model meets the fine-tuning requirements, any of the terminal nodes is removed from the online model fine-tuning node group to obtain the updated online model fine-tuning node group.

[0163] In one of the embodiments, the model fine-tuning characteristic information includes at least one of the following information: the identification of the model parameter configuration file, computational complexity, storage complexity, implementation function, task type, expected performance gain, fine-tuning delay, and security level; the edge intelligence-based online model fine-tuning device also includes an update module, which is specifically used to update the subscribed model parameter configuration file according to the online model monitoring information to obtain the updated model parameter configuration file, and to update the model fine-tuning characteristic information to obtain the updated model fine-tuning characteristic information.

[0164] In one of the embodiments, the edge intelligence-based online model fine-tuning device also includes a return module, which is specifically used to return to the step of broadcasting model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to the model fine-tuning configuration file list until the model fine-tuning termination condition is met; wherein the model fine-tuning termination condition includes that each of the terminal nodes is moved out of the online model fine-tuning node group or receives an online model fine-tuning termination signaling issued by a network node.

[0165] In one embodiment, the acquisition module 530 is specifically used to generate model fine-tuning requirement information of a cell group based on the model fine-tuning requirement information of each terminal node to which the edge node belongs; upload the model fine-tuning requirement information of the cell group to a network node, so that the network node generates a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group; and obtain the model fine-tuning configuration file list sent by the network node.

[0166] Each module in the above-mentioned online model fine-tuning device based on edge intelligence can be implemented in whole or in part by software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0167] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store model data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an online model fine-tuning method based on edge intelligence is implemented.

[0168] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0169] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0170] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0171] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0172] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0173] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0174] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0175] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An online model fine-tuning method based on edge intelligence, characterized in that: The method is applied to an edge node of a model fine-tuning system, wherein the model fine-tuning system further includes a terminal node and a network node; and the method includes: Receiving an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, and generating an online model fine-tuning node group according to at least one of the terminal nodes; broadcasting model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to the model fine-tuning configuration file list; the model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each of the terminal nodes; Obtaining a subscription request signaling sent by any of the terminal nodes; the subscription request signaling is sent by any of the terminal nodes when a subscribed model parameter configuration file is determined based on the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling; The subscribed model parameter configuration file is transmitted to any of the terminal nodes, so that any of the terminal nodes performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

2. The method according to claim 1, characterized in that: After obtaining the updated model, the method further includes: Obtaining online model monitoring information sent by any of the terminal nodes; the online model monitoring information includes a model verification result of online verification of the updated model by any of the terminal nodes; When the online model monitoring information indicates that the updated model meets the fine-tuning requirement condition, any terminal node is removed from the online model fine-tuning node group to obtain the updated online model fine-tuning node group.

3. The method according to claim 2, characterized in that The model fine-tuning feature information includes at least one of the identification of the model parameter configuration file, computational complexity, storage complexity, implementation function, task type, expected performance gain, fine-tuning delay, and security level; After obtaining the updated online model fine-tuning node group, the method further includes: According to the online model monitoring information, the subscribed model parameter configuration file is updated to obtain the updated model parameter configuration file, and the model fine-tuning feature information is updated to obtain the updated model fine-tuning feature information.

4. The method according to claim 3, characterized in that After obtaining the updated model fine-tuning feature information, the method further includes: Returning to the step of broadcasting model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to the model fine-tuning configuration file list, until a model fine-tuning termination condition is met; The model fine-tuning termination condition includes that each of the terminal nodes is moved out of the online model fine-tuning node group or receives an online model fine-tuning termination signaling sent by the network node.

5. The method according to claim 1, characterized in that Before receiving the online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, the method further includes: Generate model fine-tuning requirement information of the cell group according to model fine-tuning requirement information of each terminal node to which the edge node belongs; Uploading the model fine-tuning requirement information of the cell group to the network node, so that the network node generates a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group; Obtain the model fine-tuning configuration file list sent by the network node.

6. An online model fine-tuning system based on edge intelligence, characterized in that: The system is used to perform the steps of the online model fine-tuning method based on edge intelligence as described in any one of claims 1 to 5; the system includes a network node, an edge node and a terminal node; The network node is used to generate a corresponding model fine-tuning configuration file list based on the pre-trained model according to the model fine-tuning requirement information of the cell group uploaded by the edge node; The model fine-tuning requirement information of the cell group is generated according to the model fine-tuning requirement information of each terminal node to which the edge node belongs; The edge node is used to obtain the model fine-tuning configuration file list sent by the network node; The terminal node is used to send an online model fine-tuning request to the edge node; The edge node is configured to generate an online model fine-tuning node group according to at least one of the terminal nodes, and broadcast a model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to the model fine-tuning configuration file list; The terminal node is used to determine the subscribed model parameter configuration file according to the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling, and send a subscription request signaling to the edge node; The edge node is used to transmit the subscribed model parameter configuration file to the terminal node; The terminal node is used to perform online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

7. An online model fine-tuning device based on edge intelligence, characterized in that: An edge node applied to a model fine-tuning system, wherein the model fine-tuning system further includes a terminal node and a network node; the device includes: A receiving module, configured to receive an online model fine-tuning request uploaded by at least one terminal node to which the edge node belongs, and generate an online model fine-tuning node group according to at least one of the terminal nodes; A broadcast module, configured to broadcast model fine-tuning feature signaling to each of the terminal nodes in the online model fine-tuning node group according to a model fine-tuning configuration file list; the model fine-tuning configuration file list includes at least one model fine-tuning feature information generated according to the model fine-tuning requirement information of each of the terminal nodes; An acquisition module, used to acquire a subscription request signaling sent by any of the terminal nodes; the subscription request signaling is sent by any of the terminal nodes when a subscribed model parameter configuration file is determined based on the node operation information in the current time period and the model fine-tuning feature information carried by the model fine-tuning feature signaling; A transmission module is used to transmit the subscribed model parameter configuration file to any of the terminal nodes, so that any of the terminal nodes performs online model fine-tuning on the pre-trained model based on the subscribed model parameter configuration file to obtain an updated model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images