Information determination method and apparatus, and device, storage medium and program product

Abstract

Disclosed in the embodiments of the present disclosure are an information determination method and apparatus, and a device, a computer storage medium and a computer program product. The method comprises: sending a first request to a first network element or a second network element, wherein the first request comprises an identifier of a third network element and a selection criterion, and is used for acquiring information that can be used for model training; on the basis of the identifier of the third network element, receiving first feedback information sent by the first network element or the second network element, wherein the first feedback information comprises an identifier of a fourth network element, and is determined on the basis of the selection criterion; on the basis of the identifier of the fourth network element, sending a model training request to a target network element; and receiving second feedback information sent by the target network element.

Description

Information determination method and apparatus, device, storage medium, and program product

[0001] Cross-reference to Related Applications

[0002] The present disclosure claims priority from Chinese Patent Application No. 202410870446.3 filed on June 28, 2024 in China, the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] The present disclosure relates to information determination techniques in the field of communications, and in particular to an information determination method, apparatus, device, computer storage medium and computer program product. BACKGROUND

[0004] In an open radio access network (O-RAN) system, the functions that can reflect intelligence mainly include two modules, namely, a non real time radio intelligent controller (Non-RT RIC) and a near real time radio intelligent controller (Near-RT RIC). In the Non-RT RIC, a process is defined in which an artificial intelligence (AI) / machine language (ML) training service consumer requests a model training service from an AI / ML training service producer. Then, the training service consumer can interact with a service management and exposure function (SME) network element and a model management and exposure function (MME) network element to obtain training service producer information that can provide the training service, and obtain feedback information of model training from the training service producer. However, there are multiple training service producers that can provide the training service for the training service consumer. At this time, the SME network element will feed back multiple training service producer information, and the training service consumer can need to make multiple service requests to obtain training feedback, resulting in large signaling consumption, and even failing to obtain the training feedback. SUMMARY

[0005] To solve the above technical problems, the present disclosure provides an information determination method, device, equipment, computer storage medium and computer program product, which solves the problem of large signaling consumption in the feedback information of model training obtained by a training service consumer in the related art, and ensures that the training service consumer can obtain the feedback information of model training.

[0006] To achieve the above purpose, the technical scheme of the present disclosure is implemented as follows:

[0007] An information determination method, the method comprising:

[0008] sending a first request to a first network element or a second network element; wherein the first request comprises an identifier of a third network element and a selection criterion; the first request is used to obtain information capable of model training;

[0009] receiving first feedback information sent by the first network element or the second network element based on the identifier of the third network element; wherein the first feedback information comprises an identifier of a fourth network element; the first feedback information is determined based on the selection criterion;

[0010] based on the identifier of the fourth network element, sending a model training request to a target network element;

[0011] receiving second feedback information sent by the target network element.

[0012] In the above scheme, based on the identifier of the fourth network element, sending a model training request to a target network element, comprising:

[0013] determining a target identifier of the target network element from a plurality of identifiers of the fourth network element based on model training capability information; wherein the first feedback information further comprises at least the model training capability information;

[0014] based on the target identifier, sending the model training request to the target network element.

[0015] In the above scheme, the selection criterion comprises one or more of the following:

[0016] an identifier of a model training service;

[0017] model demand information related to a model;

[0018] an identifier of a model;

[0019] a demand priority;

[0020] a desired number of fourth network elements for feedback;

[0021] Correspondingly, the model demand information related to a model comprises one or more of the following:

[0022] resource requirement of the model, algorithm requirement of the model, input requirement of the model, output requirement of the model, training framework requirement of the model, software library requirement of the model, and scenario requirement of the model.

[0023] In the scheme, the model training capability information comprises one or more of the following:

[0024] model information corresponding to the model requirement information related to the model;

[0025] an identifier of the model;

[0026] the model information.

[0027] In the scheme, the model information comprises application scenario of the model and / or data information of the model.

[0028] Correspondingly, the model information corresponding to the model requirement information related to the model comprises one or more of the following:

[0029] resource information of the model, algorithm information of the model, input information of the model, output information of the model, training framework information of the model, and software library information of the model.

[0030] A method for determining information, the method comprising:

[0031] receiving a first request sent by a third network element; wherein the first request comprises an identifier of the third network element and a selection criterion; and the first request is used for obtaining information capable of model training;

[0032] authorizing the third network element in response to the first request;

[0033] if model training capability information synchronized by a second network element is received, sending first feedback information to the third network element corresponding to the identifier of the third network element; wherein the first feedback information comprises an identifier of a fourth network element; and the first feedback information is determined based on the selection criterion.

[0034] In the scheme, the method further comprises:

[0035] if the model training capability information synchronized by the second network element is not received, sending a second request to the second network element; wherein the second request comprises the identifier of the third network element, the selection criterion, and a uniform resource identifier.

[0036] In the scheme, the method further comprises:

[0037] if current model training capability information sent by the second network element is received, determining a first time corresponding to the current model training capability information;

[0038] If an interval between the first time and a time corresponding to the first request is less than a target time threshold, it is determined that the model training capability information of the second network element synchronization is received.

[0039] In the above scheme, the method further comprises:

[0040] If a new fourth network element is added, a query address request sent by the second network element is received.

[0041] In response to the query address request, a uniform resource identifier of the new fourth network element is sent to the second network element.

[0042] An information determination method, the method comprising:

[0043] A first request sent by a third network element is received or a second request sent by a first network element is received; wherein the first request comprises an identification of the third network element and a selection criterion; the second request comprises the identification of the third network element, the selection criterion and a uniform resource identifier; the first request is used to obtain information capable of model training;

[0044] In response to the first request or the second request, first feedback information is sent to the third network element corresponding to the identification of the third network element; wherein the first feedback information comprises an identification of a fourth network element; the first feedback information is determined based on the selection criterion.

[0045] In the above scheme, in response to the first request or the second request, the first feedback information is sent to the third network element corresponding to the identification of the third network element, comprising:

[0046] In response to the first request or the second request, model training capability information is searched, and the identification of the fourth network element is determined based on the selection criterion;

[0047] The first feedback information is sent to the third network element corresponding to the identification of the third network element; wherein the first feedback information further comprises the model training capability information.

[0048] In the above scheme, the method further comprises:

[0049] If a new fourth network element is added, and the identification of the new fourth network element is determined based on the selection criterion;

[0050] Third feedback information is sent to the third network element corresponding to the identification of the third network element; wherein the third feedback information comprises one or more of the identification of the new fourth network element, training information, the model training capability information and a uniform resource identifier.

[0051] Before the sending the first feedback information to the third network element corresponding to the identifier of the third network element, the method further includes:

[0052] sending a third request to the first network element; wherein the third request comprises the identifier of the third network element and the identifier of the fourth network element;

[0053] receiving fourth feedback information sent by the first network element; wherein the fourth feedback information comprises one or more of the identifier of the fourth network element, training information and a uniform resource identifier.

[0054] Before the sending the first feedback information to the third network element corresponding to the identifier of the third network element, the method further includes:

[0055] sending the first request to the first network element, so that the first network element sends fifth feedback information to the third network element; wherein the fifth feedback information comprises one or more of the identifier of the fourth network element, training information and a uniform resource identifier.

[0056] In the method, the training information comprises one or more of:

[0057] an identifier of a model training service;

[0058] an identifier registered by a model training service;

[0059] an identifier of a model;

[0060] model information.

[0061] A first information determination apparatus comprises:

[0062] a first sending unit configured to send a first request to a first network element or a second network element; wherein the first request comprises an identifier of a third network element and a selection criterion; the first request is used to obtain information capable of model training;

[0063] a first receiving unit configured to receive first feedback information sent by the first network element or the second network element based on the identifier of the third network element; wherein the first feedback information comprises an identifier of a fourth network element; the first feedback information is determined based on the selection criterion;

[0064] the first sending unit is further configured to send a model training request to a target network element based on the identifier of the fourth network element;

[0065] the first receiving unit is further configured to receive second feedback information sent by the target network element.

[0066] A second information determination apparatus comprises:

[0067] The second receiving unit is configured to receive a first request sent by a third network element; wherein the first request comprises an identifier of the third network element and a selection criterion; and the first request is used to acquire information capable of model training;

[0068] The first processing unit is configured to authorize the third network element in response to the first request.

[0069] The second sending unit is configured to send first feedback information to the third network element corresponding to the identifier of the third network element if the model training capability information synchronized by the second network element is received; wherein the first feedback information comprises an identifier of a fourth network element; and the first feedback information is determined based on the selection criterion.

[0070] A third information determination apparatus comprises:

[0071] The third receiving unit is configured to receive a first request sent by a third network element or receive a second request sent by a first network element; wherein the first request comprises an identifier of the third network element and a selection criterion; the second request comprises the identifier of the third network element, the selection criterion and a uniform resource identifier; and the first request is used to acquire information capable of model training.

[0072] The third sending unit is configured to send first feedback information to the third network element corresponding to the identifier of the third network element in response to the first request or the second request; wherein the first feedback information comprises an identifier of a fourth network element; and the first feedback information is determined based on the selection criterion.

[0073] A third network element comprises a first processor, a first memory and a first communication bus.

[0074] The first communication bus is configured to realize communication connection between the first processor and the first memory.

[0075] The first processor is configured to execute an information determination program in the first memory to realize the steps of the information determination method.

[0076] A first network element comprises a second processor, a second memory and a second communication bus.

[0077] The second communication bus is configured to realize communication connection between the second processor and the second memory.

[0078] The second processor is configured to execute an information determination program in the second memory to realize the steps of the information determination method.

[0079] A second network element comprises a third processor, a third memory and a third communication bus.

[0080] The third communication bus is configured to realize communication connection between the third processor and the third memory.

[0081] The third processor is configured to execute an information determination program in the third memory to realize the steps of the information determination method.

[0082] A computer readable storage medium stores one or more programs, which can be executed by one or more processors to realize the steps of the information determination method.

[0083] A computer program product includes a computer program, which, when executed by a processor, realizes the method.

[0084] The information determination method, device, equipment, computer storage medium and computer program product provided by the present disclosure send a first request for obtaining information capable of model training to a first network element or a second network element, the first request includes an identifier of a third network element and a selection criterion, receive first feedback information including an identifier of a fourth network element sent by the first network element or the second network element based on the identifier of the third network element, the first feedback information is determined based on the selection criterion, send a model training request to a target network element based on the identifier of the fourth network element, and then receive second feedback information sent by the target network element. In this way, after the training service consumer interacts with the SME network element and the MME network element to obtain the identifiers of multiple training service producers, a target training service producer is selected from the multiple training service producers, and the training service consumer can obtain feedback information of model training from the target training service producer without multiple service requests. The problem of large signaling consumption when the training service consumer obtains feedback information of model training in the related art is solved, and it is ensured that the training service consumer can obtain feedback information of model training. BRIEF DESCRIPTION OF DRAWINGS

[0085] FIG. 1 is a flowchart of an information determination method provided by an embodiment of the present disclosure;

[0086] FIG. 2 is a flowchart of another information determination method provided by an embodiment of the present disclosure;

[0087] FIG. 3 is a flowchart of still another information determination method provided by an embodiment of the present disclosure;

[0088] FIG. 4 is a flowchart of an information determination method provided by another embodiment of the present disclosure;

[0089] FIG. 5 is a flowchart of another information determination method provided by another embodiment of the present disclosure;

[0090] FIG. 6 is a flow diagram of another information determination method according to another embodiment of the present disclosure;

[0091] FIG. 7 is a flow diagram of an information determination method according to another embodiment of the present disclosure;

[0092] FIG. 8 is a structural diagram of a first information determination apparatus according to an embodiment of the present disclosure;

[0093] FIG. 9 is a structural diagram of a second information determination apparatus according to an embodiment of the present disclosure;

[0094] FIG. 10 is a structural diagram of a third information determination apparatus according to an embodiment of the present disclosure;

[0095] FIG. 11 is a structural diagram of a third network element according to an embodiment of the present disclosure;

[0096] FIG. 12 is a structural diagram of a first network element according to an embodiment of the present disclosure;

[0097] FIG. 13 is a structural diagram of a second network element according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0098] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure.

[0099] It should be understood that the "embodiments of the present disclosure" or "the foregoing embodiments" mentioned throughout the specification mean that the specific features, structures or characteristics related to the embodiments are included in at least one embodiment of the present disclosure. Therefore, "in the embodiments of the present disclosure" or "in the foregoing embodiments" appearing throughout the specification do not necessarily mean the same embodiments. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. In various embodiments of the present disclosure, the size of the serial number of each process does not mean the order of execution. The execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure. The serial number of the above-mentioned embodiments of the present disclosure is only for description, not representing the advantages and disadvantages of the embodiments.

[0100] Unless otherwise specified, the electronic device executes any step in the embodiments of the present disclosure, which can be a processor of the electronic device executing the step. It is also worth noting that the embodiments of the present disclosure do not limit the order of the steps executed by the electronic device. In addition, the way data is processed in different embodiments can be the same method or different method. It should be noted that any step in the embodiments of the present disclosure can be independently executed by the electronic device, that is, the electronic device can execute any step in the embodiments below without depending on the execution of other steps.

[0101] It should be understood that the specific embodiments described herein are merely illustrative of the present disclosure and do not limit the present disclosure.

[0102] The embodiments of the present disclosure provide an information determination method, which can be applied in a third network element. Referring to FIG. 1, the method can include the following steps:

[0103] In step 101, a first request is sent to a first network element or a second network element.

[0104] The first request includes an identification of the third network element and a selection criterion; and the first request is used to obtain information capable of model training.

[0105] In the embodiments of the present disclosure, the first network element can refer to an SME network element, a network repository function (NRF) network element, an access and mobility management function (AMF) network element, or an authentication server function (AUSF) network element, etc., and the second network element can refer to an MME network element or an AI / ML workflow function network element; the third network element can refer to a training service consumer; the identification of the third network element can refer to a training service consumer identifier, which is used to uniquely identify the training service consumer; and the first request can refer to a request for querying or discovering a model training service. In a feasible implementation manner, the identification of the third network element can be a training service consumer identification (ID). It should be noted that the present disclosure can be applied in an O-RAN system; the model training service is registered in the SME network element, and the model training capability information is registered in the AI / ML workflow function network element or the MME network element.

[0106] In step 102, first feedback information sent by the first network element or the second network element is received based on the identification of the third network element.

[0107] The first feedback information includes an identification of a fourth network element; and the first feedback information is determined based on the selection criterion.

[0108] In the embodiments of the present disclosure, after the training service consumer sends the first request to the SME network element and the MME network element, the training service consumer can receive the first feedback information returned by the SME network element and the MME network element; the first feedback information is determined by the SME network element and the MME network element based on the selection criterion, and is used to inform the training service consumer of the identification of the available training service producer having the model training capability.

[0109] The fourth network element can refer to a training service producer, and the identifier of the fourth network element can refer to a training service producer identifier. The first feedback information can refer to a request feedback of querying or discovering a model training service.

[0110] Step 103: Based on the identifier of the fourth network element, a model training request is sent to a target network element.

[0111] In the embodiments of the present disclosure, the target network element can refer to a network element determined from the fourth network element corresponding to the identifier of the fourth network element. That is, the training service consumer can determine a target producer in the training service producer according to the received training service producer identifier, and send a model training request to the target producer.

[0112] Step 104: Second feedback information sent by the target network element is received.

[0113] In the embodiments of the present disclosure, the second feedback information is feedback information returned by the target producer for the model training request.

[0114] The information determination method provided by the embodiments of the present disclosure can be used to solve the problem of large signaling consumption in the related art that the training service consumer obtains feedback information of model training, and ensure that the training service consumer can obtain feedback information of model training.

[0115] Based on the foregoing embodiments, the embodiments of the present disclosure provide an information determination method, which can be applied to a first network element. Referring to FIG. 2, the method can include the following steps:

[0116] Step 201: A first request sent by a third network element is received.

[0117] The first request includes an identifier of the third network element and a selection criterion. The first request is used to obtain information capable of model training.

[0118] Step 202: The third network element is authorized in response to the first request.

[0119] In the embodiments of the present disclosure, after the SME network element (or other first network element) receives the first request sent by the training service consumer, the training service consumer can be authorized.

[0120] Step 203: If the model training capability information synchronized by the second network element is received, first feedback information is sent to the third network element corresponding to the identifier of the third network element.

[0121] The first feedback information includes an identifier of the fourth network element; and the first feedback information is determined based on the selection criteria.

[0122] In the embodiments of the present disclosure, after the SME network element authorizes the training service consumer, and after determining that the model training capability information of the MME network element is received, the SME network element can send the first feedback information determined based on the selection criteria to the training service consumer; in this way, the training service consumer can obtain the second feedback information for the model training request from the target producer determined based on the identifier of the training service producer.

[0123] The information determination method provided by the embodiments of the present disclosure can be used in the second network element, and can include the following steps with reference to FIG. 3:

[0124] Based on the foregoing embodiments, the embodiments of the present disclosure provide an information determination method, which can be applied to a second network element. With reference to FIG. 3, the method can include the following steps:

[0125] Step 301: receiving a first request sent by a third network element or receiving a second request sent by a first network element.

[0126] The first request includes an identifier of the third network element and selection criteria; the second request includes the identifier of the third network element, the selection criteria, and a uniform resource identifier; and the first request is used to obtain information capable of performing model training.

[0127] In the embodiments of the present disclosure, the MME network element receives a request for querying or discovering model training services sent by the training service consumer; or the MME network element receives a request for querying or discovering model training capability (information) sent by the SME network element.

[0128] The second request can further include an identifier of a fourth network element and / or training information, and the uniform resource identifier can refer to a uniform resource identifier (URI).

[0129] Step 302: in response to the first request or the second request, sending first feedback information to the third network element corresponding to the identifier of the third network element.

[0130] The first feedback information includes an identifier of the fourth network element; and the first feedback information is determined based on the selection criteria.

[0131] In the embodiment of the present disclosure, after it is determined that the first request is received, the first feedback information determined based on the selection criteria can be sent to the training service consumer; in this way, the training service consumer can obtain the second feedback information for the model training request from the target producer determined according to the identifier of the training service producer.

[0132] The information determination method provided by the embodiment of the present disclosure can enable the training service consumer to select a target training service producer after interacting with the SME network element and the MME network element to obtain the identifiers of a plurality of training service producers, so that the training service consumer can obtain the feedback information of model training from the target training service producer without multiple service requests, thereby solving the problem of large signaling consumption in the related art when the training service consumer obtains the feedback information of model training, and ensuring that the training service consumer can obtain the feedback information of model training.

[0133] Based on the foregoing embodiments, the embodiment of the present disclosure provides an information determination method, which can include the following steps with reference to FIGS. 4-7:

[0134] Step 401: The third network element sends a first request to the first network element or the second network element.

[0135] The first request includes an identifier of the third network element and selection criteria; and the first request is used to obtain information capable of model training.

[0136] In the embodiment of the present disclosure, the selection criteria include one or more of the following:

[0137] An identifier of a model training service;

[0138] Model demand information related to a model;

[0139] An identifier of a model;

[0140] A demand priority;

[0141] A fourth network element quantity expected to be fed back;

[0142] In the embodiment of the present disclosure, the model demand information related to a model includes one or more of the following:

[0143] Resource demand of a model, algorithm demand of a model, input demand of a model, output demand of a model, training framework demand of a model, software library demand of a model, and scenario demand of a model.

[0144] In other embodiments of the present disclosure, the algorithm demand of a model can be specifically described as shown in Table 1, and the input demand of a model can be specifically described as shown in Table 2.

[0145] Table 1

[0146] Table 2

[0147] Wherein, PRB refers to Physical Resource Block, and RRC refers to Radio Resource Control.

[0148] It should be noted that after step 401, steps 402-406 can be optionally executed, or steps 407-412 can be executed.

[0149] Step 402, the second network element receives the first request sent by the third network element.

[0150] It should be noted that after step 402, steps 403-405 can be executed as shown in FIG. 4, or step 406 can be executed as shown in FIG. 5.

[0151] Step 403, the second network element sends a third request to the first network element.

[0152] Wherein, the third request includes the identification of the third network element and the identification of the fourth network element.

[0153] In the embodiments of the present disclosure, the third request can refer to a request for querying or discovering a model training service; specifically, the MME network element also sends a request for querying or discovering a model training service to the SME network element.

[0154] Step 404, the second network element receives the fourth feedback information sent by the first network element.

[0155] Wherein, the fourth feedback information includes one or more of the identification of the fourth network element, the training information, and the uniform resource identifier.

[0156] In the embodiments of the present disclosure, after the SME network element receives the request for querying or discovering a model training service sent by the MME network element, the SME network element can respond to the request for querying or discovering a model training service and send a request feedback (i.e., the fourth feedback information) to the MME network element.

[0157] 405, the second network element sends the first feedback information to the third network element corresponding to the identification of the third network element.

[0158] Wherein, the first feedback information includes the identification of the fourth network element; the first feedback information is determined based on the selection criteria.

[0159] In the embodiments of the present disclosure, after receiving the fourth feedback information sent by the SME network element, the MME network element can send the first feedback information determined based on the selection criteria and including the identifier of the fourth network element to the training service consumer.

[0160] 406. The second network element sends a first request to the first network element, so that the first network element sends fifth feedback information to the third network element.

[0161] The fifth feedback information includes one or more of the identifier of the fourth network element, the training information, and the uniform resource identifier.

[0162] In the embodiments of the present disclosure, after receiving the first request sent by the training service consumer, the MME network element also sends the first request sent by the training service consumer to the SME network element, so that the SME network element directly feeds back the request (i.e., the fifth feedback information) of querying or discovering the model training service to the training service consumer.

[0163] Step 407, the first network element receives the first request sent by the third network element.

[0164] Step 408, the first network element authorizes the third network element in response to the first request.

[0165] It should be noted that after step 408 as shown in FIG. 6, steps 409-410 can be optionally executed; or after step 408 as shown in FIG. 7, steps 411-412 are executed.

[0166] Step 409, if the current model training capability information sent by the second network element is received, the first network element determines a first time corresponding to the current model training capability information.

[0167] In the embodiments of the present disclosure, an information synchronization mechanism can be set between the SME network element and the MME network element. Every n time interval, if the model training capability information is updated, the MME network element sends the current model training capability information to the SME network element, otherwise, it is not sent. Wherein, n can be a time interval pre-set according to specific application scenarios and requirements.

[0168] It should be noted that the first time can refer to the time when the SME network element sends the current model training capability information, or the time when the MME network element receives the current model training capability information.

[0169] Step 410, if the interval between the first time and a second time corresponding to the first request is less than a target time threshold, the first network element determines that the model training capability information synchronized by the second network element is received, and sends the first feedback information to the third network element corresponding to the identifier of the third network element.

[0170] In the embodiments of the present disclosure, the SME network element only records one synchronization timestamp (i.e., the first time) corresponding to the current model training capability information; the second time refers to the time when the training service consumer sends the first request; when the time interval between the second time of the first request of the training service consumer and the latest synchronization timestamp (i.e., the first time) is less than a target time threshold, the model training capability information of the SME network element and the MME network element is considered to be synchronized, and then the SME network element sends the first feedback information to the corresponding training service consumer. The target time threshold can be a time length set in advance according to specific application scenarios and requirements.

[0171] Step 411: If the model training capability information synchronized by the second network element is not received, the first network element sends a second request to the second network element.

[0172] The second request includes the identifier of the third network element, the selection criteria, and a uniform resource identifier.

[0173] Step 412: The second network element receives the second request sent by the first network element, and sends the first feedback information to the third network element corresponding to the identifier of the third network element in response to the second request.

[0174] It should be noted that the "sending the first feedback information to the third network element corresponding to the identifier of the third network element" in steps 405 and 412 can be implemented in the following way:

[0175] A1: The second network element finds the model training capability information, and determines the identifier of the fourth network element based on the selection criteria.

[0176] A2: The second network element sends the first feedback information to the third network element corresponding to the identifier of the third network element.

[0177] The first feedback information further includes the model training capability information.

[0178] It should be noted that the first feedback information can further include one or more of the following: model identifier, training information, notification URI.

[0179] It should be noted that steps 413 can be performed after steps 405, 410 and 412 as shown in FIGS. 4, 6 and 7, and step 414 can be performed after step 406 as shown in FIG. 5. In one possible implementation, the fourth network element in FIGS. 4-7 can refer to a target network element determined from the fourth network element.

[0180] Step 413: The third network element receives the first feedback information sent by the first network element or the second network element.

[0181] Step 414: The third network element receives the fifth feedback information sent by the first network element.

[0182] It should be noted that after step 413 and step 414, step 415 can be executed.

[0183] Step 415, the third network element determines the target identifier of the target network element from the identifiers of the plurality of fourth network elements based on the model training capability information.

[0184] The model training capability information includes one or more of the following:

[0185] Model information corresponding to the model-related model requirement information;

[0186] An identifier of the model;

[0187] Model information.

[0188] The model information includes an application scenario of the model and / or data information of the model.

[0189] The model information corresponding to the model-related model requirement information includes one or more of the following:

[0190] Resource information of the model, algorithm information of the model, input information of the model, output information of the model, training framework information of the model, and software library information of the model.

[0191] It should be noted that the algorithm information of the model and the input information of the model are respectively the responses to the algorithm requirement of the model and the input requirement of the model; the data information of the model includes but is not limited to one or more of the following: data acquisition information, data time length, data source, model precision, and data size (here, the data information is different from the response to the input requirement, and is the training data information of the existing model).

[0192] Step 416, the third network element sends a model training request to the target network element based on the target identifier.

[0193] In the embodiments of the present disclosure, if the training service consumer obtains identifiers of a plurality of training service producers, the training service consumer sorts the training service producers providing the obtained model training capability information and the self requirement according to the sorting result, and the training service consumer can select an identifier of a training service producer with a high ranking as the target identifier based on the sorting result, and send a request for a model training service to a target producer corresponding to the target identifier.

[0194] In other embodiments of the present disclosure, the method can further include:

[0195] Step 417, if a new fourth network element is added, the first network element receives a query address request sent by the second network element.

[0196] The query address request can be a request for querying the address of the training service producer.

[0197] Step 418, the first network element sends a uniform resource identifier of the new fourth network element to the second network element in response to the query address request.

[0198] Wherein, the SME network element can feed back the URI of the newly added new training service producer to the MME network element in response to the query address request.

[0199] Step 419, the second network element sends third feedback information to the third network element corresponding to the identification of the third network element based on the selection criteria.

[0200] Wherein, the third feedback information includes one or more of the identification of the new fourth network element, training information, model training capability information and uniform resource identifier.

[0201] In the embodiments of the present disclosure, the MME network element obtains the URI of the newly added new training service producer, and after selecting the final new training service producer based on the selection criteria, can send an update producer notification (i.e. the third feedback information) to the training service consumer.

[0202] In other embodiments of the present disclosure, the training information includes one or more of the following:

[0203] The identification of the model training service;

[0204] The identification of the model training service registration;

[0205] The identification of the model;

[0206] The model information.

[0207] It should be noted that the number of interactions between the training service consumer and the platform in the present disclosure is significantly reduced; and the SME network element does not need to feed back the training information to the training service consumer; the SME network element and the MME network element coordinate the training service consumer demand and the training service producer information within the platform, and there is only one outlet for the training service consumer, avoiding information difference; at the same time, it avoids the request conflict of multiple training service consumers to the same training service producer. At the same time, the linkage feedback of the SME network element and the MME network element, the MME network element can intelligently feed back the training service producer information according to the demand of different training service consumers at this moment, avoiding subsequent service conflict and further reducing communication overhead.

[0208] Further, the training service consumer adds a query or discovers a (unique or multiple) training service producer capable of this model training before requesting model training, thereby ensuring the execution of the training service and reducing unnecessary communication overhead (fewer steps, fewer interactions, and less information sent) due to the inability to identify a qualified training service producer.

[0209] In other embodiments of the present disclosure, the training service consumer can send a query model training capability information request to the MME network element; the query model training capability information request includes the training service consumer ID and the selection criteria; the MME network element authorizes the training service consumer and sends a query model training capability information feedback to the training service consumer, wherein the query model training capability information feedback includes the training service producer ID and the model training capability information. Then, the training service consumer sends a discovery model training service request to the SME network element, and the discovery model training service request includes the training service producer ID and the training service consumer ID; at this time, the SME network element authorizes the training service consumer and sends a discovery model training service feedback to the training service consumer; wherein the discovery model training service feedback includes one or more of the training service producer ID, the URI, and the training information. Then, the training service consumer sends a model training service request to the training service producer corresponding to the training service producer ID, and the training service producer returns a model training service feedback to the training service consumer. The model training service request includes the model ID and the evaluation criteria; the model training service feedback includes the training job ID.

[0210] The information determination method provided by the embodiments of the present disclosure can be used to solve the problem of large signaling consumption in the related art when the training service consumer obtains the feedback information of model training from the target training service producer without multiple service requests, and ensure that the training service consumer can obtain the feedback information of model training.

[0211] Based on the foregoing embodiments, the embodiments of the present disclosure provide a first information determination apparatus, which can be applied to the information determination method provided by the embodiments corresponding to FIG. 1 and FIG. 4 to FIG. 7. Referring to FIG. 8, the first information determination apparatus 5 can include a first sending unit 51 and a first receiving unit 52, wherein:

[0212] The first sending unit 51 is configured to send a first request to the first network element or the second network element, wherein the first request comprises an identifier of the third network element and a selection criterion; and the first request is used to obtain information capable of model training.

[0213] The first receiving unit 52 is configured to receive first feedback information sent by the first network element or the second network element based on the identifier of the third network element, wherein the first feedback information comprises an identifier of a fourth network element; and the first feedback information is determined based on the selection criterion.

[0214] The first sending unit 51 is further configured to send a model training request to a target network element based on the identifier of the fourth network element.

[0215] In other embodiments of the present disclosure, the first sending unit 51 is further configured to perform the following steps:

[0216] determining a target identifier of a target network element from the identifiers of the plurality of fourth network elements based on the model training capability information; wherein the first feedback information further comprises the model training capability information;

[0217] sending the model training request to the target network element based on the target identifier.

[0218] In other embodiments of the present disclosure, the selection criterion comprises one or more of the following:

[0219] an identifier of a model training service;

[0220] model demand information related to a model;

[0221] an identifier of a model;

[0222] a demand priority;

[0223] a number of fourth network elements expected to provide feedback.

[0224] In other embodiments of the present disclosure, the model demand information related to the model comprises one or more of the following:

[0225] a resource demand of the model, an algorithm demand of the model, an input demand of the model, an output demand of the model, a training framework demand of the model, a software library demand of the model, and a scenario demand of the model.

[0226] In other embodiments of the present disclosure, the model training capability information comprises one or more of the following:

[0227] model information corresponding to the model demand information related to the model;

[0228] an identifier of a model;

[0229] model information.

[0230] In other embodiments of the present disclosure, the model information includes an application scenario of the model and / or data information of the model.

[0231] In other embodiments of the present disclosure, the model information corresponding to the model requirement information related to the model includes one or more of the following:

[0232] The resource information of the model, the algorithm information of the model, the input information of the model, the output information of the model, the training framework information of the model, and the software library information of the model.

[0233] It should be noted that the specific implementation process of the steps performed by each unit in the embodiments of the present disclosure can refer to the implementation process in the information determination method provided in the corresponding embodiments of FIGS. 1 and 4-7, which will not be described here.

[0234] The first information determination apparatus provided by the embodiments of the present disclosure can select a target training service producer from a plurality of training service producers after the training service consumer interacts with the SME network element and the MME network element to obtain the identifiers of the plurality of training service producers. The training service consumer can obtain the feedback information of the model training from the target training service producer without multiple service requests. The problem of large signaling consumption in obtaining the feedback information of the model training by the training service consumer in the related art is solved, and the training service consumer can obtain the feedback information of the model training.

[0235] Based on the foregoing embodiments, the embodiments of the present disclosure provide a second information determination apparatus, which can be applied to the information determination method provided in the corresponding embodiments of FIGS. 2 and 4-7. Referring to FIG. 9, the second information determination apparatus 6 can include a second receiving unit 61, a first processing unit 62, and a second sending unit 63, wherein:

[0236] The second receiving unit 61 is configured to receive a first request sent by a third network element, wherein the first request includes an identifier of the third network element and a selection criterion; and the first request is used to obtain information capable of model training.

[0237] The first processing unit 62 is configured to authorize the third network element in response to the first request.

[0238] The second sending unit 63 is configured to send first feedback information to the third network element corresponding to the identifier of the third network element if the model training capability information synchronized by the second network element is received, wherein the first feedback information includes an identifier of a fourth network element; and the first feedback information is determined based on the selection criterion.

[0239] In other embodiments of the present disclosure, the first processing unit 62 is further configured to send a second request to the second network element if the model training capability information synchronized by the second network element is not received, wherein the second request comprises the identifier of the third network element, the selection criteria and the uniform resource identifier.

[0240] In other embodiments of the present disclosure, the first processing unit 62 is further configured to perform the following steps:

[0241] If the current model training capability information sent by the second network element is received, a first time corresponding to the current model training capability information is determined.

[0242] If the interval between the first time and the time corresponding to the first request is less than the target time threshold, it is determined that the model training capability information synchronized by the second network element is received.

[0243] In other embodiments of the present disclosure, the second receiving unit 61 is further configured to receive a query address request sent by the second network element if a new fourth network element is added.

[0244] The second sending unit 63 is further configured to send the uniform resource identifier of the new fourth network element to the second network element in response to the query address request.

[0245] It should be noted that the specific implementation process of the steps performed by each unit in the embodiments of the present disclosure can refer to the implementation process in the information determination method provided in the corresponding embodiments of FIGS. 2 and 4-7, which will not be described here.

[0246] The second information determination apparatus provided by the embodiments of the present disclosure can be used to solve the problem of large signaling consumption in the related art when the training service consumer obtains the feedback information of model training from the target training service producer selected from the plurality of training service producers after interacting with the SME network element and the MME network element to obtain the identifiers of the plurality of training service producers, and ensure that the training service consumer can obtain the feedback information of model training.

[0247] Based on the foregoing embodiments, the embodiments of the present disclosure provide a third information determination apparatus, which can be applied to the information determination method provided in the corresponding embodiments of FIGS. 3-7. Referring to FIG. 10, the third information determination apparatus 7 can comprise a third receiving unit 71 and a third sending unit 72, wherein:

[0248] The third receiving unit 71 is configured to receive a first request sent by a third network element or receive a second request sent by a first network element, wherein the first request comprises an identifier of the third network element and a selection criterion; the second request comprises the identifier of the third network element, the selection criterion, and a uniform resource identifier; and the first request is used to obtain information capable of model training.

[0249] The third sending unit 72 is configured to send first feedback information to the third network element corresponding to the identifier of the third network element in response to the first request or the second request, wherein the first feedback information comprises an identifier of a fourth network element; and the first feedback information is determined based on the selection criterion.

[0250] In other embodiments of the present disclosure, the third sending unit 72 is further configured to perform the following steps:

[0251] In response to the first request or the second request, the model training capability information is searched, and the identifier of the fourth network element is determined based on the selection criterion.

[0252] The first feedback information is sent to the third network element corresponding to the identifier of the third network element; wherein the first feedback information further comprises the model training capability information.

[0253] In other embodiments of the present disclosure, the third sending unit 72 is further configured to, if a new fourth network element is added and the identifier of the new fourth network element is determined based on the selection criterion, send third feedback information to the third network element corresponding to the identifier of the third network element.

[0254] The third feedback information comprises one or more of the identifier of the new fourth network element, the training information, the model training capability information, and the uniform resource identifier.

[0255] In other embodiments of the present disclosure, the third sending unit 72 is further configured to send a third request to the first network element, wherein the third request comprises the identifier of the third network element and the identifier of the fourth network element.

[0256] The third receiving unit 71 is further configured to receive fourth feedback information sent by the first network element, wherein the fourth feedback information comprises one or more of the identifier of the fourth network element, the training information, and the uniform resource identifier.

[0257] In other embodiments of the present disclosure, the third sending unit 72 is further configured to send the first request to the first network element, so that the first network element sends fifth feedback information to the third network element, wherein the fifth feedback information comprises one or more of the identifier of the fourth network element, the training information, and the uniform resource identifier.

[0258] In other embodiments of the present disclosure, the training information comprises one or more of the following:

[0259] An identifier of a model training service;

[0260] an identity of the model training service registration;

[0261] an identity of the model;

[0262] model information.

[0263] It should be noted that the specific implementation process of the steps performed by each unit in the embodiments of the present disclosure can be determined with reference to the implementation process in the information determination method provided in the corresponding embodiments of FIGS. 3-7, which will not be described here.

[0264] The third information determination apparatus provided by the embodiments of the present disclosure can select a target training service producer from a plurality of training service producers after the training service consumer interacts with the SME network element and the MME network element to obtain the identities of the plurality of training service producers, and the training service consumer can obtain the feedback information of the model training from the target training service producer without multiple service requests, thereby solving the problem of large signaling consumption in obtaining the feedback information of the model training by the training service consumer in the related art, and ensuring that the training service consumer can obtain the feedback information of the model training.

[0265] Based on the foregoing embodiments, the embodiments of the present disclosure provide a third network element, which can be applied to the information determination method provided in the corresponding embodiments of FIGS. 1 and 4-7. Referring to FIG. 11, the third network element 8 can include a first processor 81, a first memory 82, and a first communication bus 83, wherein:

[0266] The first communication bus 83 is configured to realize the communication connection between the first processor 81 and the first memory 82;

[0267] The first processor 81 is configured to execute an information determination program in the first memory 82 to realize the following steps:

[0268] send a first request to the first network element or the second network element; wherein the first request includes an identity of the third network element and a selection criterion; the first request is used to obtain information capable of model training;

[0269] receive first feedback information sent by the first network element or the second network element based on the identity of the third network element; wherein the first feedback information includes an identity of the fourth network element; the first feedback information is determined based on the selection criterion;

[0270] send a model training request to a target network element based on the identity of the fourth network element.

[0271] In other embodiments of the present disclosure, the first processor 81 is configured to execute the information determination program in the first memory 82 to send a model training request to a target network element based on the identity of the fourth network element, to realize the following steps:

[0272] determine the target identifier of the target network element from the identifiers of the plurality of fourth network elements based on the model training capability information; wherein the first feedback information further comprises at least the model training capability information;

[0273] send a model training request to the target network element based on the target identifier.

[0274] In other embodiments of the present disclosure, the selection criteria include one or more of the following:

[0275] an identifier of the model training service;

[0276] model requirement information related to the model;

[0277] an identifier of the model;

[0278] a demand priority;

[0279] a number of fourth network elements expected to provide feedback.

[0280] In other embodiments of the present disclosure, the model requirement information related to the model includes one or more of the following:

[0281] a resource requirement of the model, an algorithm requirement of the model, an input requirement of the model, an output requirement of the model, a training framework requirement of the model, a software library requirement of the model, and a scenario requirement of the model.

[0282] In other embodiments of the present disclosure, the model training capability information includes one or more of the following:

[0283] model information corresponding to the model requirement information related to the model;

[0284] an identifier of the model;

[0285] model information.

[0286] In other embodiments of the present disclosure, the model information includes an application scenario of the model and / or data information of the model.

[0287] In other embodiments of the present disclosure, the model information corresponding to the model requirement information related to the model includes one or more of the following:

[0288] resource information of the model, algorithm information of the model, input information of the model, output information of the model, training framework information of the model, and software library information of the model.

[0289] It should be noted that the specific description of the steps performed by the first processor can refer to the implementation process in the information determination method provided in the corresponding embodiments of FIGS. 1 and 4-7, which will not be repeated here.

[0290] The third network element provided by the embodiments of the present disclosure can be used to solve the problem of large signaling consumption in the related art when the training service consumer obtains the feedback information of model training, and ensure that the training service consumer can obtain the feedback information of model training.

[0291] Based on the foregoing embodiments, the embodiments of the present disclosure provide a first network element, which can be applied to the information determination method provided by the embodiments corresponding to FIG. 2 and FIG. 4 to FIG. 7. Referring to FIG. 12, the first network element 9 can include a second processor 91, a second memory 92 and a second communication bus 93, wherein:

[0292] The second communication bus 93 is configured to realize the communication connection between the second processor 91 and the second memory 92;

[0293] The second processor 91 is configured to execute an information determination program in the second memory 92 to realize the following steps:

[0294] receive a first request sent by a third network element; wherein the first request includes an identification of the third network element and a selection criterion; the first request is used to obtain information capable of model training;

[0295] authorize the third network element in response to the first request;

[0296] if the model training capability information synchronized by the second network element is received, send first feedback information to the third network element corresponding to the identification of the third network element; wherein the first feedback information includes an identification of a fourth network element; the first feedback information is determined based on the selection criterion.

[0297] In other embodiments of the present disclosure, the second processor 91 is configured to execute the information determination program in the second memory 92, and can also realize the following steps:

[0298] if the model training capability information synchronized by the second network element is not received, send a second request to the second network element; wherein the second request includes the identification of the third network element, the selection criterion and a uniform resource identifier.

[0299] In other embodiments of the present disclosure, the second processor 91 is configured to execute the information determination program in the second memory 92, and can also realize the following steps:

[0300] if the current model training capability information sent by the second network element is received, determine a first time corresponding to the current model training capability information;

[0301] If an interval between the first time and a time corresponding to the first request is less than a target time threshold, it is determined that the model training capability information of the second network element synchronization is received.

[0302] In other embodiments of the present disclosure, the second processor 91 is configured to execute an information determination program in the second memory 92, and the following steps can also be implemented:

[0303] If a new fourth network element is added, a query address request sent by the second network element is received.

[0304] In response to the query address request, a uniform resource identifier of the new fourth network element is sent to the second network element.

[0305] It should be noted that the specific description of the steps performed by the second processor can refer to the implementation process of the information determination method provided in the corresponding embodiments of FIGS. 2 and 4-7, which will not be described here.

[0306] The first network element provided by the embodiments of the present disclosure can interact with the SME network element and the MME network element to obtain the identifiers of a plurality of training service producers after the training service consumer, and then select a target training service producer. The training service consumer can obtain the feedback information of model training from the target training service producer without multiple service requests, thereby solving the problem of large signaling consumption in obtaining the feedback information of model training by the training service consumer in the related art, and ensuring that the training service consumer can obtain the feedback information of model training.

[0307] Based on the foregoing embodiments, the embodiments of the present disclosure provide a second network element, which can be applied to the information determination method provided in the corresponding embodiments of FIGS. 3-7. Referring to FIG. 13, the second network element 10 can include a third processor 1001, a third memory 1002, and a third communication bus 1003, wherein:

[0308] The third communication bus 1003 is configured to realize the communication connection between the third processor 1001 and the third memory 1002;

[0309] The third processor 1001 is configured to execute an information determination program in the third memory 1002 to implement the following steps:

[0310] Receive a first request sent by a third network element or receive a second request sent by a first network element; wherein the first request includes an identifier of the third network element and a selection criterion; the second request includes the identifier of the third network element, the selection criterion, and a uniform resource identifier; and the first request is used to obtain information capable of model training;

[0311] In response to the first request or the second request, the first feedback information is sent to the third network element corresponding to the identifier of the third network element; wherein the first feedback information comprises the identifier of the fourth network element; and the first feedback information is determined based on the selection standard.

[0312] In other embodiments of the present disclosure, the third processor 1001 is configured to execute the information determination program in the third memory 1002 to, in response to the first request or the second request, send the first feedback information to the third network element corresponding to the identifier of the third network element, so as to implement the following steps:

[0313] In response to the first request or the second request, the model training capability information is searched, and the identifier of the fourth network element is determined based on the selection standard;

[0314] The first feedback information is sent to the third network element corresponding to the identifier of the third network element; wherein the first feedback information further comprises the model training capability information.

[0315] In other embodiments of the present disclosure, the third processor 1001 is configured to execute the information determination program in the third memory 1002, and can also implement the following steps:

[0316] If a new fourth network element is added and the identifier of the new fourth network element is determined based on the selection standard, the third feedback information is sent to the third network element corresponding to the identifier of the third network element;

[0317] The third feedback information comprises one or more of the identifier of the new fourth network element, the training information, the model training capability information and the uniform resource identifier.

[0318] In other embodiments of the present disclosure, the third processor 1001 is configured to execute the information determination program in the third memory 1002, and can also implement the following steps:

[0319] The third request is sent to the first network element; wherein the third request comprises the identifier of the third network element and the identifier of the fourth network element;

[0320] The fourth feedback information sent by the first network element is received; wherein the fourth feedback information comprises one or more of the identifier of the fourth network element, the training information and the uniform resource identifier.

[0321] In other embodiments of the present disclosure, the third processor 1001 is configured to execute the information determination program in the third memory 1002, and can also implement the following steps:

[0322] The first request is sent to the first network element, so that the first network element sends the fifth feedback information to the third network element; wherein the fifth feedback information comprises one or more of the identifier of the fourth network element, the training information and the uniform resource identifier.

[0323] In other embodiments of the present disclosure, the training information comprises one or more of the following:

[0324] an identity of the model training service;

[0325] an identity registered by the model training service;

[0326] an identity of the model;

[0327] model information.

[0328] It should be noted that the specific description of the steps performed by the third processor can refer to the implementation process in the information determination method provided by the corresponding embodiments of FIGS. 3-7, which will not be repeated here.

[0329] According to the second network element provided by the embodiments of the present disclosure, after the training service consumer interacts with the SME network element and the MME network element to obtain the identities of a plurality of training service producers, the training service consumer selects a target training service producer from the plurality of training service producers, and can obtain the feedback information of the model training from the target training service producer without multiple service requests, thereby solving the problem of large signaling consumption in the related art when the training service consumer obtains the feedback information of the model training, and ensuring that the training service consumer can obtain the feedback information of the model training.

[0330] Based on the foregoing embodiments, the embodiments of the present disclosure provide a computer-readable storage medium, which stores one or more programs executable by one or more processors to implement the steps in the information determination method provided by the corresponding embodiments of FIGS. 1-7.

[0331] Based on the foregoing embodiments, the embodiments of the present disclosure provide a computer program product, which includes a computer program executable by the first processor 81, the second processor 91, and the third processor 1001 to complete the steps of the information determination method provided by the corresponding embodiments of FIGS. 1-7.

[0332] Those skilled in the art should understand that the embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure can be in the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer usable program code.

[0333] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0334] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks.

[0335] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0336] The above description is only specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present disclosure, which should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A method for determining information, comprising: Send a first request to a first network element or a second network element; wherein the first request includes the identifier and selection criteria of a third network element; the first request is used to obtain information that enables model training; The first feedback information sent by the first network element or the second network element is received based on the identifier of the third network element; wherein, the first feedback information includes the identifier of the fourth network element; the first feedback information is determined based on the selection criteria; Based on the identifier of the fourth network element, a model training request is sent to the target network element; Receive the second feedback information sent by the target network element.

2. The method according to claim 1, wherein, Sending a model training request to the target network element based on the identifier of the fourth network element includes: Based on model training capability information, the target identifier of the target network element is determined from the identifiers of multiple fourth network elements; wherein, the first feedback information further includes at least the model training capability information; Based on the target identifier, the model training request is sent to the target network element.

3. The method according to claim 1, wherein, The selection criteria include one or more of the following: Identifier for model training service; Model-related requirements information; Model identifier; Prioritizing requirements; The expected number of fourth network elements to be fed back; Accordingly, the model-related model requirements information includes one or more of the following: The model's resource requirements, algorithm requirements, input requirements, output requirements, training framework requirements, software library requirements, and scenario requirements.

4. The method according to claim 2, wherein, The model training capability information includes one or more of the following: Model information corresponding to model-related model requirements; Model identifier; Model information.

5. The method according to claim 4, wherein, The model information includes the application scenarios of the model and / or the data information of the model; Accordingly, the model information corresponding to the model-related model requirement information includes one or more of the following: Model resource information, model algorithm information, model input information, model output information, model training framework information, and model software library information.

6. An information determination method, comprising: Receive a first request sent by a third network element; wherein the first request includes the identifier and selection criteria of the third network element; the first request is used to obtain information that enables model training; Authorize the third network element in response to the first request; If the model training capability information synchronized by the second network element is received, a first feedback message is sent to the third network element corresponding to the identifier of the third network element; wherein, the first feedback message includes the identifier of the fourth network element; the first feedback message is determined based on the selection criteria.

7. The method according to claim 6, further comprising: If the model training capability information synchronized by the second network element is not received, a second request is sent to the second network element; wherein, the second request includes the identifier, selection criteria and uniform resource identifier of the third network element.

8. The method according to claim 6, further comprising: If the current model training capability information sent by the second network element is received, the first time corresponding to the current model training capability information is determined; If the interval between the first time and the time corresponding to the first request is less than the target time threshold, it is determined that the model training capability information synchronized by the second network element has been received.

9. The method according to claim 6, further comprising: If a new fourth network element is added, it receives the address query request sent by the second network element; In response to the query address request, the Uniform Resource Identifier of the new fourth network element is sent to the second network element.

10. An information determination method, comprising: The system receives a first request from a third network element or a second request from a first network element; wherein the first request includes the identifier and selection criteria of the third network element; the second request includes the identifier, selection criteria, and Uniform Resource Identifier of the third network element; and the first request is used to obtain information that enables model training. In response to the first request or the second request, a first feedback message is sent to the third network element corresponding to the identifier of the third network element; wherein, the first feedback message includes the identifier of the fourth network element; the first feedback message is determined based on the selection criteria.

11. The method according to claim 10, wherein, The step of responding to the first request or the second request by sending first feedback information to the third network element corresponding to the identifier of the third network element includes: In response to the first request or the second request, retrieve model training capability information and determine the identifier of the fourth network element based on the selection criteria; The first feedback information is sent to the third network element corresponding to the identifier of the third network element; wherein, the first feedback information also includes the model training capability information.

12. The method of claim 10, further comprising: If a new fourth network element is added and its identifier is determined based on the selection criteria, a third feedback message is sent to the third network element corresponding to its identifier; wherein the third feedback message includes one or more of the following: the identifier of the new fourth network element, training information, model training capability information, and Uniform Resource Identifier.

13. The method according to claim 10, wherein, Before sending the first feedback information to the third network element corresponding to the identifier of the third network element, the method further includes: Send a third request to the first network element; wherein the third request includes the identifier of the third network element and the identifier of the fourth network element; The system receives a fourth feedback message sent by the first network element; wherein the fourth feedback message includes one or more of the following: the identifier of the fourth network element, training information, and a Uniform Resource Identifier.

14. The method of claim 10, wherein, Before sending the first feedback information to the third network element corresponding to the identifier of the third network element, the method further includes: The first request is sent to the first network element so that the first network element sends the fifth feedback information to the third network element; wherein the fifth feedback information includes one or more of the identifier of the fourth network element, training information, and uniform resource identifier.

15. The method according to claim 12, wherein, The training information includes one or more of the following: Identifier for model training service; The identifier for model training service registration; Model identifier; Model information.

16. A first information determining device, comprising: The first sending unit is configured to send a first request to a first network element or a second network element; wherein the first request includes the identifier and selection criteria of a third network element; the first request is used to obtain information that enables model training. The first receiving unit is configured to receive first feedback information sent by the first network element or the second network element based on the identifier of the third network element; wherein the first feedback information includes the identifier of the fourth network element; and the first feedback information is determined based on the selection criteria. The first sending unit is also used to send a model training request to the target network element based on the identifier of the fourth network element; The first receiving unit is also used to receive second feedback information sent by the target network element.

17. A second information determining device, comprising: The second receiving unit is used to receive a first request sent by a third network element; wherein the first request includes the identifier and selection criteria of the third network element; the first request is used to obtain information that enables model training. The first processing unit is configured to respond to the first request to authorize the third network element; The second sending unit is configured to send first feedback information to the third network element corresponding to the identifier of the third network element if it receives model training capability information synchronized by the second network element; wherein the first feedback information includes the identifier of the fourth network element; the first feedback information is determined based on the selection criteria.

18. A third information determining device, comprising: The third receiving unit is configured to receive a first request sent by a third network element or a second request sent by a first network element; wherein the first request includes the identifier and selection criteria of the third network element; the second request includes the identifier, selection criteria, and Uniform Resource Identifier of the third network element; the first request is used to obtain information that enables model training. The third sending unit is configured to respond to the first request or the second request by sending first feedback information to the third network element corresponding to the identifier of the third network element; wherein, the first feedback information includes the identifier of the fourth network element; the first feedback information is determined based on the selection criteria.

19. A third network element, comprising: A first processor, a first memory, and a first communication bus; The first communication bus is used to establish a communication connection between the first processor and the first memory; The first processor is configured to execute an information determination program in the first memory to implement the steps of the information determination method as described in any one of claims 1 to 5.

20. A first network element, comprising: Second processor, second memory, and second communication bus; The second communication bus is used to establish a communication connection between the second processor and the second memory; The second processor is used to execute the information determination program in the second memory to implement the steps of the information determination method as described in any one of claims 6 to 9.

21. A second network element, comprising: A third processor, a third memory, and a third communication bus; The third communication bus is used to realize the communication connection between the third processor and the third memory; The third processor is used to execute the information determination program in the third memory to implement the steps of the information determination method as described in any one of claims 10 to 15.

22. A computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the steps of the information determination method as described in any one of claims 1-5, 6-9, or 10-15.

23. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 5, 6 to 9, or 10 to 15.

Images