Operation related to AI / ML models

Abstract

The exemplary embodiments of this disclosure relate to operations related to artificial intelligence / machine learning (AI / ML) models. In one exemplary method, a first network device sends a first request to a second network device indicating to the second network device that it should provide a first AI / ML model, and receives the first AI / ML model from the second network device. The first network device then obtains a finely tuned AI / ML model based on the first AI / ML model. In this way, a customized local model can be obtained from a global base model to reduce the complexity of training at random access network (RAN) nodes. At radio access network (RAN) nodes, a relatively lightweight customized AI / ML model can be obtained from a fairly large global base model in the core network (CN) or a third party, reducing the complexity of training at the RAN node.

Claims

1. The steps include: a first network device sending a first request to a second network device instructing the second network device to provide a first artificial intelligence / machine learning (AI / ML) model; The step of receiving the first AI / ML model from the second network device; The step of obtaining a finely tuned AI / ML model based on the first AI / ML model described above. Methods that include...

2. Obtaining the aforementioned fine-tuned AI / ML model involves: This includes performing fine-tuning on the first AI / ML model based on data from the first network device, The method according to claim 1.

3. The steps include sending a second request to the second network device instructing the second network device to provide a second AI / ML model; The step of receiving the second AI / ML model from the second network device and The method according to claim 2, further comprising:

4. The method according to claim 3, wherein the second request is transmitted together with the first request, and the second AI / ML model is received together with the first AI / ML model.

5. Receiving the second AI / ML model together with the first AI / ML model means: At least one model parameter common to the first AI / ML model and the second AI / ML model; At least one model parameter specific to the first AI / ML model described above; At least one model parameter specific to the second AI / ML model described above and The method according to claim 4, which includes receiving

6. The steps include: transmitting the finely tuned AI / ML model to the second network device; The step of receiving the updated AI / ML model from the second network device mentioned above. The method according to any one of claims 1 to 5, further comprising:

7. The steps include: transmitting the finely tuned AI / ML model to at least one terminal device; The steps include receiving at least one third AI / ML model from the aforementioned at least one terminal device; A step of generating an updated AI / ML model based on the aforementioned at least one third AI / ML model. The method according to any one of claims 1 to 6, further comprising:

8. The at least one terminal device includes multiple terminal devices, the at least one third AI / ML model includes multiple AI / ML models, and the updated AI / ML model is generated based on the at least one third AI / ML model: This includes aggregating the aforementioned multiple AI / ML models to generate a fourth AI / ML model as the updated AI / ML model. The method according to claim 7.

9. Generating the updated AI / ML model based on the at least one AI / ML model is: The method further includes aggregating the fourth AI / ML model and the fine-tuned AI / ML model to generate a fifth AI / ML model as the updated AI / ML model. The method according to claim 8.

10. The steps include: transmitting the updated AI / ML model to the second network device; The step of receiving an updated AI / ML model from the second network device mentioned above. The method according to any one of claims 7 to 9, further comprising:

11. The process further includes the step of transmitting the data to the second network device before receiving the AI / ML model from the second network device, The AI / ML model received by the first network device is a finely tuned AI / ML model that has been fine-tuned based on the data. The method according to claim 2.

12. A step of performing the task using at least one of the AI / ML model, the fine-tuned AI / ML model, the updated AI / ML model, or the further updated AI / ML model, A step relating to at least one of the aforementioned tasks, of transmitting data stored in the AI / ML database on the first network device to the second network device. The method according to any one of claims 1 to 11, further comprising at least one of the above.

13. The second network device receives a first request from the first network device instructing the second network device to provide an artificial intelligence / machine learning (AI / ML) model; The second network device comprises the step of generating an AI / ML model based on a pre-trained AI / ML model, wherein the pre-trained AI / ML model is pre-trained for a plurality of tasks, including tasks to be performed by the first network device using the AI / ML model; The step of transmitting the AI / ML model to the first network device, method.

14. The steps include: receiving a second request from the first network device instructing the second network device to provide a second AI / ML model; The steps include: generating a second AI / ML model based on the aforementioned pre-trained AI / ML model; The step of transmitting the second AI / ML model to the first network device. The method according to claim 13, further comprising:

15. The method according to claim 14, wherein the second request is received together with the first request, and the second AI / ML model is transmitted together with the first AI / ML model.

16. Sending the second AI / ML model together with the first AI / ML model is: At least one model parameter common to the first AI / ML model and the second AI / ML model; At least one model parameter specific to the first AI / ML model described above; At least one model parameter specific to the second AI / ML model described above and The method according to claim 15, which includes transmitting

17. The steps include: receiving at least one AI / ML model, including an AI / ML model provided by the first network device, from at least one network device, including the first network device; The steps include: generating an updated AI / ML model for performing the task based on the at least one AI / ML model; The steps include transmitting the updated AI / ML model to the first network device. The method according to any one of claims 13 to 16, further comprising:

18. A step of receiving at least one AI / ML model from at least one network device, including the first network device, which includes an updated AI / ML model provided by the first network device, wherein the updated AI / ML model is generated based on at least one AI / ML model provided by at least one terminal device; A step of generating an updated AI / ML model based on the aforementioned at least one AI / ML model; The steps include transmitting the updated AI / ML model to the first network device. The method according to any one of claims 13 to 17, further comprising:

19. The method according to claim 13, wherein the AI / ML model transmitted to the first network device is a finely tuned AI / ML model based on data from the first network device.

20. Before transmitting the AI / ML model to the first network device, the AI / ML model is to receive data from the first network device for fine-tuning the AI / ML model; The steps include: performing fine-tuning on the AI / ML model based on the received data and obtaining the fine-tuned AI / ML model; and The method according to claim 19, further comprising:

21. The steps include: receiving data stored in the first network device from the first network device in relation to at least one of the aforementioned tasks; The step of storing the received data in the second network device The method according to any one of claims 13 to 20, further comprising:

22. The terminal device receives an artificial intelligence / machine learning (AI / ML) model from a first network device; The steps include: performing fine-tuning on the AI / ML model based on the data collected by the terminal device and obtaining an updated, fine-tuned AI / ML model; The steps include transmitting the updated AI / ML model to the first network device and Methods that include...

23. Transceiver and; A processor that is communicatively coupled to the aforementioned transceiver A first network device having, The aforementioned processor is: A first request is transmitted to a second network device via the transceiver, instructing the second network device to provide a first artificial intelligence / machine learning (AI / ML) model; The first AI / ML model is received from the second network device via the transceiver; Based on the first AI / ML model described above, obtain a finely tuned AI / ML model. A first network device configured as follows.

24. Transceiver and; A processor that is communicatively coupled to the aforementioned transceiver A second network device having, The aforementioned processor is: The steps include: receiving a first request from a first network device via the transceiver instructing the second network device to provide an artificial intelligence / machine learning (AI / ML) model; The second network device comprises the step of generating an AI / ML model based on a pre-trained AI / ML model, wherein the pre-trained AI / ML model is pre-trained for a plurality of tasks, including tasks to be performed by the first network device using the AI / ML model; The steps include transmitting the AI / ML model to the first network device via the transceiver, and A second network device configured to perform the following actions.

25. Transceiver and; A processor that is communicatively coupled to the aforementioned transceiver A terminal device having the following processor: The transceiver receives an artificial intelligence / machine learning (AI / ML) model from the first network device; Based on the data collected by the terminal device, fine-tune the AI / ML model and obtain an updated AI / ML model; The transceiver is configured to transmit the updated AI / ML model to the first network device. Terminal device.

26. A non-temporary computer-readable medium storing a computer program, wherein the computer program, when executed on at least one processor, causes the at least one processor to execute the method according to any one of claims 1 to 22.

27. A chip having at least one processing circuit configured to perform the method described in any one of claims 1 to 22.

28. A computer program product having a computer-executable instruction that, when stored in tangible form on a computer-readable medium, causes a device to perform the method according to any one of claims 1 to 22.

Images