Background data transfer policy for the ai / ML enablement client re-selection

Abstract

Systems and methods are disclosed for provisioning and management of Artificial Intelligence (AI) / Machine Learning (ML) data transfer policy for AI / ML member re-selection. In one embodiment, a method performed by an AI / ML enabler server comprises receiving, from a Vertical Application Layer (VAL) server, an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member re-selection and storing the AI / ML data transfer policy for AI / ML member re-selection. In this manner, a mechanism for provisioning and management of an AI / ML data transfer policy for AI / ML member re-selection is provided.

Description

BACKGROUND DATA TRANSFER POLICY FOR THE AI / ML ENABLEMENT CLIENT RE-SELECTIONRelated Applications

[0001] This application claims the benefit of International Patent Application No.PCT / CN 2024 / 086209, filed April 5, 2024, the disclosure of which is hereby incorporated herein by reference in its entirety.Technical Field

[0002] The present disclosure relates to a wireless communication system (e.g., a 3rdGeneration Partnership Project (3GPP) system) and, more specifically, Artificial Intelligence (Al) / Machine Learning (ML) policy provisioning and management.Background

[0003] Third Generation Partnership Project (3GPP) Technical Report (TR) 23.700-82VO.3.0 states:Data analytics is a useful tool for the operator to help optimizing the service offering by predicting events related to the network or slice or UE conditions.3GPP introduced data analytics function (NWDAF) [2] to support network data analytics services in 5G Core network, management data analytics service (MDAS) [3] to provide data analytics at the 0AM, and application data analytics service (ADAES) [4].In this direction, the support for AI / ML services in 3GPP system has been studied for providing AI / ML enabled analytics in NWDAF, as well as for assisting the ASP / 3rd party AI / ML application service provider for the AI / ML model distribution, transfer, training for various applications (e.g., video / speech recognition, robot control, automotive).Considering vertical-specific applications and edge applications as the major consumers of 3GPP-provided data analytics services, the application enablement layer can play role on the exposure of AI / ML services from different 3GPP domains to the vertical / ASP in a unified manner; and on defining, at an overarching layer, value-add support services for assisting AI / ML services provided by either the VAL layer or the application enablement layer (for enhancing the SEAL ADAES services).

[0004] 3GPP TR 23.700-82 identifies the key issues and corresponding application architecture and related solutions. More specifically, 3GPP TR 23.700-82 identifies the application enabling layer architecture, capabilities, and services to support Artificial Intelligence (AI) / Machine Learning (ML) services at the application layer. The application-layer AI / ML services shall support multiple operations (e.g., regular,distributed, vertical / horizontal learning) and operate with multiple heterogeneous members that can participate or be selected for participation in the AI / ML operations.

[0005] Systems and methods are disclosed for provisioning and management of Artificial Intelligence (Al) I Machine Learning (ML) data transfer policy for AI / ML member re-selection. In one embodiment, a method performed by an AI / ML enabler server comprises receiving, from a Vertical Application Layer (VAL) server, an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member re-selection and storing the AI / ML data transfer policy for AI / ML member re-selection. In this manner, a mechanism for provisioning and management of an AI / ML data transfer policy for AI / ML member re-selection is provided.

[0006] In one embodiment, the method further comprises applying the AI / ML data transfer policy for AI / ML member re-selection.

[0007] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection is a AI / ML data transfer policy to be applied for AI / ML data transfer from a dis- selected AI / ML member to a selected AI / ML member during an AI / ML member reselection procedure.

[0008] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection comprises information that indicates a policy action to be taken. In one embodiment, the policy action is initiating background AI / ML data transfer. In another embodiment, the policy action is configuring a temporary AI / ML data transfer with Quality of Service (QoS) during the AI / ML member re-selection procedure.

[0009] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection comprises a list of required AI / ML traffic Quality of Service, QoS, values that represent required QoS values that are to be enforced for AI / ML data transfer with QoS.

[0010] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection comprises background data transmission details and policy guidance. In one embodiment, the background data transmission details and policy guidance comprises: a VAL service identifier of a VAL service to which the request applies, a list of VAL User Equipment, UE, identifiers for which the policy applies or a VAL group identifier, an expected data volume for the background data transfer, and a desired geographic area for the background data transfer. In one embodiment, the background datatransmission details and policy guidance further comprises an expiration time for the background transfer and / or policy selection guidance with respect to selection from multiple transfer policies provided by an underlying network.

[0011] In one embodiment, the method further comprises sending a response to the VAL server.

[0012] Corresponding embodiments of an AI / ML enabler server and a network node for implementing an AI / ML enabler server are also disclosed.

[0013] Embodiments of a method performed by a VAL server are also disclosed. In one embodiment, a method performed by a VAL server comprises sending, to an AI / ML enabler server, an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member re-selection.

[0014] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection is a AI / ML data transfer policy to be applied for AI / ML data transfer from a dis- selected AI / ML member to a selected AI / ML member during an AI / ML member reselection procedure.

[0015] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection comprises information that indicates a policy action to be taken. In one embodiment, the policy action is initiating background AI / ML data transfer. In another embodiment, the policy action is configuring a temporary AI / ML data transfer with QoS during the AI / ML member re-selection procedure.

[0016] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection comprises a list of required AI / ML traffic Quality of Service, QoS, values that represent required QoS values that are to be enforced for AI / ML data transfer with QoS.

[0017] In one embodiment, the AI / ML data transfer policy for AI / ML member reselection comprises background data transmission details and policy guidance. In one embodiment, the background data transmission details and policy guidance comprises: a VAL service identifier of a VAL service to which the request applies, a list of VAL User Equipment, UE, identifiers for which the policy applies or a VAL group identifier, an expected data volume for the background data transfer, and a desired geographic area for the background data transfer. In one embodiment, the background data transmission details and policy guidance further comprises an expiration time for the background transfer and / or policy selection guidance with respect to selection from multiple transfer policies provided by an underlying network.

[0018] Corresponding embodiments of a VAL server and a network node for implementing a VAL server are also disclosed.Brief Description of the Drawing Figures

[0019] The accompanying drawing figures incorporated in and forming a part of this specification illustrate several aspects of the disclosure, and together with the description serve to explain the principles of the disclosure.

[0020] Figure 1 illustrates a procedure for provisioning and management of Artificial Intelligence (Al) I Machine Learning (ML) data transfer policy for AI / ML member reselection, in accordance with one embodiment of the present disclosure; and

[0021] Figures 2 and 3 illustrate example embodiments of a network node in which aspects of embodiments of the present disclosure may be implemented.Detailed Description

[0022] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Other embodiments, however, are contained within the scope of the subject matter disclosed herein, the disclosed subject matter should not be construed as limited to only the embodiments set forth herein; rather, these embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0023] Third Generation Partnership Project (3GPP) Technical Report (TR) 23.700-82 considers the Artificial Intelligence (AI) / Machine Learning (ML) client re-selection procedure based on the provided AI / ML policies. During the AI / ML Enablement Client re-selection procedure, the AI / ML data transfer may be needed, e.g., the current state of the trained ML model from client A shall be transferred to client B.

[0024] There currently exist certain problems. The existing solution does not support the AI / ML data transfer configuration during AI / ML Enablement Client re-selection.Thus, the policy for the AI / ML data transfer during the AI / ML member re-selection should be added in order to provide an application-friendly experience.

[0025] Certain aspects of the present disclosure and their embodiments may provide solutions to the aforementioned or other challenges. Embodiments of the present disclosure relate to an application layer AI / ML policy for AI / ML data transfer duringAI / ML Enablement Client re-selection. In one embodiment, the application layer AI / ML policy for AI / ML data transfer during AI / ML Enablement Client re-selection includes:- Policy ID: The Policy ID is an identifier that can be used for the policy application and enforcement in another operations;- Policy Action: The policy action is to either initiate background AI / ML data transfer or configure a temporary AI / ML data transfer with Quality of Service (QoS) during AI / ML member re-selection procedure;- (Optional) List of one or more required QoS values that are to be enforced for the AI / ML data transfer with QoS for the "configure a temporary AI / ML data transfer with QoS during AI / ML member re-selection procedure" action;- (Optional) Background data transmission details and policy guidance for the "initiate background AI / ML data transfer" action. The background data transmission details and policy guidance may be, for example, as described in clause 14.3.2.58 of 3GPP TS 23.434 V19.1.0, which is reproduced below.*****TS23.434, Clause 14.3.2.58 *****14.3.2.58 BDT configuration requestTable 14.3.58-1 describes the information flow for the BDT configuration request from the VAL server to the NRM Server.Table 14.3.2.58-1 : BDT configuration request*****END TS23.434, Clause 14.3.2.58 *****

[0026] Certain embodiments may provide one or more of the following technical advantage(s). The following use-case examples are supported by the application layer AI / ML policy for AI / ML data transfer during AI / ML Enablement Client re-selection disclosed herein:- A Vertical Application Layer (VAL) server performs time-sensitive learning that should be completed as soon as possible. The VAL server can request the AI / ML Enablement server to configure a temporary AI / ML data transfer session with QoS during the AI / ML member re-selection procedure, in order to transfer the current trained data within a minimal amount of time.- A VAL server performs non-time-sensitive learning and would like to reduce the cost of the AI / ML operation. The VAL server can request the AI / ML Enablement server to use background data transfer with lowest cost as Background Data Transfer (BDT) policy guidance.

[0027] Figure 1 illustrate a policy provisioning and management procedure for an application layer AI / ML policy for AI / ML data transfer during AI / ML Enablement Client re-selection, in accordance with one example embodiment of the present disclosure. The procedure involves a VAL server 100 and an AI / ML enablement server 102. In this example embodiment, policy provisioning and management for this new policy is implemented via the existing procedure in 3GPP TR 23.700-82, Clause 8.14.2.1 (see step 1 and step 3 in particular).

[0028] The steps of the procedure of Figure 1 are as follows:

[0029] Step 1: The VAL server 100 sends an AI / ML service request with policies provisioning and management information as defined in Table 1 below. As shown in Table 1, in an embodiment of the present disclosure, the AI / ML service request includesan AI / ML data transfer policy for AI / ML member re-selection. In one embodiment, the AI / ML data transfer policy for AI / ML member re-selection includes the information shown in Table 4 (i.e., Policy ID, Action, an optional List of the required AI / ML traffic QoS values, and optional background data transmission details and policy guidance).

[0030] Step 2: Upon receiving the request, the AI / ML Enablement Server 102 performs an authorization check of the VAL server 100.

[0031] Step 3: If the VAL server 100 is authorized, the AI / ML Enablement Server 102 stores the AI / ML policy (or policies) included in the request of Step 1 and applies the provisioned policies to ongoing and further the AI / ML operations (e.g., ML model training).

[0032] Step 4:. AI / ML Enablement Server provides the response to the VAL server with a status via AI / ML member selection policies provisioning and management response defined in clause 8.14.3.2.

[0033] Note that the service operation defined in 3GPP TR 23.700-82 clause 8.A.2.1 can be utilized for AI / ML member selection policies management (i.e., update / delete) functionality.

[0034] Also note that the related service operation request (e.g., member selection) is to be updated in the evaluation stage.

[0035] Table 1, which is a modified version of Table 8.14.3.1-1 from 3GPP TR 23.700-82 where additions are shown via underlined text, describes the information flow from the VAL server to the AI / ML Enablement Server as a request that contains the AI / ML policies provisioning and management information.

[0036] Table 2 is a reproduction of Table 8.14.3.1-2 from 3GPP TR 23.700-82.

[0037] Table 3 is a reproduction of Table 8.14.3.1-3 from 3GPP TR 23.700-82.

[0038] Table 4 defines one example embodiment of the AI / ML data transfer policy.Table 1: AI / ML policies provisioning and management requestTable 2: Member selection and re-selection policyTable 3: AI / ML traffic QoS adjustment policyTable 4: Al / ML data transfer policy

[0039] One example implementation of an example embodiment of the present disclosure is as follows. This example implementation is described in the form of changes to 3GPP TR 23.700-82 V0.3.0.

[0040] Introduction

[0041] The re-selection procedure may require the data relocation from the dis- selected member to the newly selected member. For example:

[0042] Given: the VAL server has specific requirements for the data transfer from dis-selected member to the selected member, e.g., initiate background AI / ML datatransfer or configure a temporary AI / ML data transfer with QoS during AI / ML member re-selection procedure.

[0043] When: the re-selection conditions are met, the AI / ML Enablement Server shall configure the data transfer session with the requested characteristics and initiate the data transfer.

[0044] Then: the AI / ML process can continue on the new AI / ML client.

[0045] Reason for Change

[0046] The VAL server may have specific requirements for the AIML data transfer (e.g., the current state of the trained ML model) during the re-selection procedure.

[0047] The following use-case examples are supported by the policy:

[0048] - The VAL server performs the time-sensitive learning that should be completed as soon as possible. The VAL server can request the AI / ML Enablement server to configure a temporary AI / ML data transfer session with QoS during AI / ML member re-selection procedure, in order to transfer the current trained data within the minimal time.

[0049] - The VAL server performs the non-time-sensitive learning and would like to reduce the cost of the AI / ML operation. The VAL server can request the AI / ML Enablement server to use background data transfer with lowest costs as BDT policy guidance.

[0050] Conclusions

[0051] The following proposed changes provide a new AI / ML policy for data transfer in order to address the possible requirements from the VAL server.

[0052] Proposal

[0053] It is proposed to make the following changes to 3GPP 23.700-82 VO.3.0, where additions are show via underlined text:***** FIRST CHANGE *****8.14.3.1 AI / ML service request with policies provisioning and management informationTable 8.14.3.1-1 describes the information flow from the VAL server to the AI / ML Enablement Server as a request that contains the AI / ML policies provisioning and management information.Table 8.14.3.1-1 : AI / ML policies provisioning and management requestEditor's note: Whether additional AI / ML policies are needed is FFS.Table 8.14.3.1-2: Member selection and re-selection policyEditor's note: The AI / ML compute capability conditions are FFS.Table 8.14.3.1-3: AI / ML traffic QoS adjustment policyTable 8.14.3.1-4: AI / ML data transfer policy***** END CHANGES *****

[0054] Figure 2 is a schematic block diagram of a network node 200 according to some embodiments of the present disclosure. Optional features are represented by dashed boxes. The network node 200 may be, for example, a network node in a wireless communication system (e.g., a 5G or 6G system) such as, e.g., a RAN node or a core network node that implements the functionality of the VAL server 100 or the AI / ML Enablement Server 102 described above with respect to Figure 1. As illustrated, the network node 200 includes a control system 202 that includes one or more processors 204 (e.g., Central Processing Units (CPUs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs), and / or the like), memory 206, and a network interface 208. The one or more processors 204 are also referred to herein as processing circuitry. In addition, if the network node 200 is a radio access node (e.g., a base station, gNB, or network node that implements at least some of the functionality of the base station or gNB), the network node 200 may include one or more radio units 210 that each includes one or more transmitters 212 and one or more receivers 214 coupled to one or more antennas 216. The radio units 210 may be referred to or be part of radio interface circuitry. In some embodiments, the radio unit(s) 210 is external to the control system 202 and connected to the control system 202 via, e.g., a wired connection (e.g., an optical cable). However, in some other embodiments, the radio unit(s) 210 and potentially the antenna(s) 216 are integrated together with the control system 202. The one or more processors 204 operate to provide one or more functions of the network node 200 as described herein (e.g., one or more functions of the VAL server or the AI / ML Enabler Server described herein). In some embodiments, the function(s) are implemented in software that is stored, e.g., in the memory 206 and executed by the one or more processors 204.

[0055] Figure 3 is a schematic block diagram that illustrates a virtualized embodiment of the network node 200 according to some embodiments of the present disclosure. Again, optional features are represented by dashed boxes. As used herein, a "virtualized" network node is an implementation of the network node 200 in which at least a portion of the functionality of the network node 200 is implemented as a virtual component(s) (e.g., via a virtual machine(s) executing on a physical processing node(s) in a network(s)). As illustrated, in this example, if the network node 200 is a radio access node, the network node 200 may include the control system 202 and / or the oneor more radio units 210, as described above. The control system 202 may be connected to the radio unit(s) 210 via, for example, an optical cable or the like. The network node 200 includes one or more processing nodes 300 coupled to or included as part of a network(s) 302. If present, the control system 202 or the radio unit(s) are connected to the processing node(s) 300 via the network 302. Each processing node 300 includes one or more processors 304 (e.g., CPUs, ASICs, FPGAs, and / or the like), memory 306, and a network interface 308.

[0056] In this example, functions 310 of the network node 200 described herein (e.g., one or more functions of the VAL server 100 or the AI / ML Enablement Server 102 described herein) are implemented at the one or more processing nodes 300 or distributed across the one or more processing nodes 300 and the control system 202 and / or the radio unit(s) 210 in any desired manner. In some particular embodiments, some or all of the functions 310 of the network node 200 described herein are implemented as virtual components executed by one or more virtual machines implemented in a virtual environ ment(s) hosted by the processing node(s) 300. As will be appreciated by one of ordinary skill in the art, additional signaling or communication between the processing node(s) 300 and the control system 202 is used in order to carry out at least some of the desired functions 310. Notably, in some embodiments, the control system 202 may not be included, in which case the radio unit(s) 210 communicate directly with the processing node(s) 300 via an appropriate network interface(s).

[0057] In some embodiments, a computer program including instructions which, when executed by at least one processor, causes the at least one processor to carry out the functionality of the network node 200 or a node (e.g., a processing node 300) implementing one or more of the functions 310 of the network node 200 in a virtual environment according to any of the embodiments described herein is provided. In some embodiments, a carrier comprising the aforementioned computer program product is provided. The carrier is one of an electronic signal, an optical signal, a radio signal, or a computer readable storage medium (e.g., a non-transitory computer readable medium such as memory).

[0058] Any appropriate steps, methods, features, functions, or benefits disclosed herein may be performed through one or more functional units or modules of one or more virtual apparatuses. Each virtual apparatus may comprise a number of thesefunctional units. These functional units may be implemented via processing circuitry, which may include one or more microprocessor or microcontrollers, as well as other digital hardware, which may include Digital Signal Processor (DSPs), special-purpose digital logic, and the like. The processing circuitry may be configured to execute program code stored in memory, which may include one or several types of memory such as Read Only Memory (ROM), Random Access Memory (RAM), cache memory, flash memory devices, optical storage devices, etc. Program code stored in memory includes program instructions for executing one or more telecommunications and / or data communications protocols as well as instructions for carrying out one or more of the techniques described herein. In some implementations, the processing circuitry may be used to cause the respective functional unit to perform corresponding functions according to one or more embodiments of the present disclosure.

[0059] While processes in the figures may show a particular order of operations performed by certain embodiments of the present disclosure, it should be understood that such order is exemplary (e.g., alternative embodiments may perform the operations in a different order, combine certain operations, overlap certain operations, etc.).

Claims

Claims1. A method performed by an Artificial Intelligence, Al, I Machine Learning, ML, enablement server (102), the method comprising: receiving (Fig. 1, step 1), from a Vertical Application Layer, VAL, server (100), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member reselection; and storing (Fig. 1, step 3) the AI / ML data transfer policy for AI / ML member reselection.

2. The method of claim 1, further comprising applying (Fig. 1, step 3) the AI / ML data transfer policy for AI / ML member re-selection.

3. The method of claim 1 or 2, wherein the AI / ML data transfer policy for AI / ML member re-selection is a AI / ML data transfer policy to be applied for AI / ML data transfer from a dis-selected AI / ML member to a selected AI / ML member during an AI / ML member re-selection procedure.

4. The method of any of claims 1 to 3, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises information that indicates a policy action to be taken.

5. The method of claim 4, wherein the policy action is initiating background AI / ML data transfer.

6. The method of claim 4, wherein the policy action is configuring a temporary AI / ML data transfer with Quality of Service, QoS, during the AI / ML member re-selection procedure.

7. The method of any of claims 1 to 6, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises a list of required AI / ML traffic Quality of Service, QoS, values that represent required QoS values that are to be enforced for AI / ML data transfer with QoS.

8. The method of any of claims 1 to 7, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises background data transmission details and policy guidance.

9. The method of claim 8, wherein the background data transmission details and policy guidance comprises: a VAL service identifier of a VAL service to which the request applies, a list of VAL User Equipment, UE, identifiers for which the policy applies or a VAL group identifier, an expected data volume for the background data transfer, and a desired geographic area for the background data transfer.

10. The method of claim 9, wherein the background data transmission details and policy guidance further comprises an expiration time for the background transfer and / or policy selection guidance with respect to selection from multiple transfer policies provided by an underlying network.

11. The method of any of claims 1 to 10, further comprising sending (Fig. 1, step 4) a response to the VAL server (100).

12. An Artificial Intelligence, Al, I Machine Learning, ML, enablement server (102) adapted to: receive (Fig. 1, step 1), from a Vertical Application Layer, VAL, server (100), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member reselection; and store (Fig. 1, step 3) the AI / ML data transfer policy for AI / ML member reselection.

13. The AI / ML enablement server (102) of claim 12, further adapted to perform the method of any of claims 2 to 11.

14. A network node (200) for implementing an Artificial Intelligence, Al, I Machine Learning, ML, enablement server (102), the network node (200) comprising processing circuitry (204; 304) configured to cause the network node (200) to:receive (Fig. 1, step 1), from a Vertical Application Layer, VAL, server (100), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member reselection; and store (Fig. 1, step 3) the AI / ML data transfer policy for AI / ML member reselection.

15. The network node (200) of claim 14, wherein the processing circuitry (204; 304) is further configured to cause the network node (200) to apply (Fig. 1, step 3) the AI / ML data transfer policy for AI / ML member re-selection.

16. The network node (200) of claim 14 or 15, wherein the AI / ML data transfer policy for AI / ML member re-selection is a AI / ML data transfer policy to be applied for AI / ML data transfer from a dis-selected AI / ML member to a selected AI / ML member during an AI / ML member re-selection procedure.

17. The network node (200) of any of claims 14 to 16, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises information that indicates a policy action to be taken.

18. The network node (200) of claim 17, wherein the policy action is initiating background AI / ML data transfer.

19. The network node (200) of claim 17, wherein the policy action is configuring a temporary AI / ML data transfer with Quality of Service, QoS, during the AI / ML member re-selection procedure.

20. The network node (200) of any of claims 14 to 19, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises a list of required AI / ML traffic Quality of Service, QoS, values that represent required QoS values that are to be enforced for AI / ML data transfer with QoS.

21. The network node (200) of any of claims 14 to 20, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises background data transmission details and policy guidance.

22. The network node (200) of claim 21, wherein the background data transmission details and policy guidance comprises: a VAL service identifier of a VAL service to which the request applies, a list of VAL User Equipment, UE, identifiers for which the policy applies or a VAL group identifier, an expected data volume for the background data transfer, and a desired geographic area for the background data transfer.

23. The network node (200) of claim 22, wherein the background data transmission details and policy guidance further comprises an expiration time for the background transfer and / or policy selection guidance with respect to selection from multiple transfer policies provided by an underlying network.

24. The network node (200) of any of claims 14 to 23, wherein the processing circuitry (204; 304) is further configured to cause the network node (200) to send (Fig. 1, step 4) a response to the VAL server (100).

25. A computer program comprising instructions which, when executed on at least one processor, cause the processor to carry out the method according to any of claims 1 to 11.

26. A carrier containing the computer program of claim 25, wherein the carrier is one of an electronic signal, an optical signal, a radio signal, or a computer readable storage medium.

27. A non-transitory computer-readable medium comprising instructions executable by processing circuitry of a network node, whereby the network node is operable to: receive (Fig. 1, step 1), from a Vertical Application Layer, VAL, server (100), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member reselection; andstore (Fig. 1, step 3) the AI / ML data transfer policy for AI / ML member reselection.

28. A method performed by a Vertical Application Layer, VAL, server (100), the method comprising: sending (Fig. 1, step 1), to an Artificial Intelligence, Al, I Machine Learning, ML, enablement server (102), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member re-selection; and receiving (Fig. 1, step 4) a response from the AI / ML enablement server (102).

29. The method of claim 28, wherein the AI / ML data transfer policy for AI / ML member re-selection is a AI / ML data transfer policy to be applied for AI / ML data transfer from a dis-selected AI / ML member to a selected AI / ML member during an AI / ML member re-selection procedure.

30. The method of claim 28 or 29, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises information that indicates a policy action to be taken.

31. The method of claim 30, wherein the policy action is initiating background AI / ML data transfer.

32. The method of claim 30, wherein the policy action is configuring a temporary AI / ML data transfer with Quality of Service, QoS, during the AI / ML member re-selection procedure.

33. The method of any of claims 28 to 32, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises a list of required AI / ML traffic Quality of Service, QoS, values that represent required QoS values that are to be enforced for AI / ML data transfer with QoS.

34. The method of any of claims 28 to 33, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises background data transmission details and policy guidance.

35. The method of claim 34, wherein the background data transmission details and policy guidance comprises: a VAL service identifier of a VAL service to which the request applies, a list of VAL User Equipment, UE, identifiers for which the policy applies or a VAL group identifier, an expected data volume for the background data transfer, and a desired geographic area for the background data transfer.

36. The method of claim 35, wherein the background data transmission details and policy guidance further comprises an expiration time for the background transfer and / or policy selection guidance with respect to selection from multiple transfer policies provided by an underlying network.

37. A Vertical Application Layer, VAL, server (100) adapted to: send (Fig. 1, step 1), to an Artificial Intelligence, Al, I Machine Learning, ML, enablement server (102), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member re-selection; and receive (Fig. 1, step 4) a response from the AI / ML enablement server (102).

38. The VAL server (100) of claim 37, further adapted to perform the method of any of claims 29 to 36.

39. A network node (200) for implementing a Vertical Application Layer, VAL, server (100), the network node (200) comprising processing circuitry (204; 304) configured to cause the network node (200) to: send (Fig. 1, step 1), to an Artificial Intelligence, Al, I Machine Learning, ML, enablement server (102), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member re-selection; and receive (Fig. 1, step 4) a response from the AI / ML enablement server (102).

40. The network node of claim 39, wherein the AI / ML data transfer policy for AI / ML member re-selection is a AI / ML data transfer policy to be applied for AI / ML data transfer from a dis-selected AI / ML member to a selected AI / ML member during an AI / ML member re-selection procedure.

41. The network node of claim 39 or 40, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises information that indicates a policy action to be taken.

42. The network node of claim 41, wherein the policy action is initiating background AI / ML data transfer.

43. The network node of claim 41, wherein the policy action is configuring a temporary AI / ML data transfer with Quality of Service, QoS, during the AI / ML member re-selection procedure.

44. The network node of any of claims 39 to 43, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises a list of required AI / ML traffic Quality of Service, QoS, values that represent required QoS values that are to be enforced for AI / ML data transfer with QoS.

45. The network node of any of claims 39 to 44, wherein the AI / ML data transfer policy for AI / ML member re-selection comprises background data transmission details and policy guidance.

46. The network node of claim 45, wherein the background data transmission details and policy guidance comprises: a VAL service identifier of a VAL service to which the request applies, a list of VAL User Equipment, UE, identifiers for which the policy applies or a VAL group identifier, an expected data volume for the background data transfer, and a desired geographic area for the background data transfer.

47. The network node of claim 46, wherein the background data transmission details and policy guidance further comprises an expiration time for the background transfer and / or policy selection guidance with respect to selection from multiple transfer policies provided by an underlying network.

48. A computer program comprising instructions which, when executed on at least one processor, cause the processor to carry out the method according to any of claims 28 to 36.

49. A carrier containing the computer program of claim 48, wherein the carrier is one of an electronic signal, an optical signal, a radio signal, or a computer readable storage medium.

50. A non-transitory computer-readable medium comprising instructions executable by processing circuitry of a network node, whereby the network node is operable to: send (Fig. 1, step 1), to an Artificial Intelligence, Al, I Machine Learning, ML, enablement server (102), an AI / ML service request comprising an AI / ML data transfer policy for AI / ML member re-selection; and receive (Fig. 1, step 4) a response from the AI / ML enablement server (102).

Images