Abstract: This application provides a method and a related communication apparatus applicable to the communication field. In the method, a network device receives a first notification message sent by a terminal device, where the first notification message indicates that the terminal device has completed training on a target model, and the first notification message includes a model name corresponding to the target model. The network device then determines policy information based on the uplink channel state information of the terminal device; and sends a lightweight indication message to the terminal device.

Abstract: A model distillation training method provides for establishment of a distillation training communication connection with a second device prior to performing distillation training on a neural network model. Based on exchange of distillation training information between the first device and the second device, the second device configures a first reference neural network model by using first configuration information sent by the first device. After configuring the first reference neural network model, the second device performs operation processing on first sample data in first data information based on the configured first reference neural network model by using the first data information to obtain first indication information, and sends the first indication information to the first device. The first device trains, by using the first indication information, a first neural network model designed by the first device.

Abstract: A federated learning method, apparatus, and system are disclosed. A first node obtains data distribution information of a plurality of second nodes based on a target data feature required by a training task; the first node selects at least two target second nodes from the plurality of second nodes based on a target data class required by the training task and the data distribution information of the plurality of second nodes; and the first node indicates the at least two target second nodes to perform federated learning, to obtain a federated learning model that is in the training task and that corresponds to the target data class. In this way, when participants have a plurality of data distributions, a trained model is prevented, as much as possible, from being affected by data poisoning.

Abstract: Model evaluation methods and apparatuses are described, which may be applied to systems such as 5G and vehicle-to-everything (V2X). In an example model evaluation method, an inference device determines an inference output of a machine learning model based on an inference input of the machine learning model, where the machine learning model is determined based on first configuration information. The inference device sends a first message to a measurement device, where the first message includes a measurement object, the measurement object is determined based on the inference output, and the first message is used to request a measurement result of the measurement object. The inference device receives the measurement result from the measurement device, where the measurement result corresponds to the inference output. The inference device sends an evaluation result to an update device, where the evaluation result is determined based on the inference output and the measurement result.

Abstract: A collaborative inference method and a communication apparatus. The method includes: the terminal device determines a first inference result based on a first machine learning (ML) submodel, where the first ML submodel is a part of an ML model. The terminal device sends the first inference result. Then, the terminal device receives a target inference result, where the target inference result is an inference result that is of the ML model and that is determined based on the first inference result.

Abstract: Embodiments of this application provide a communication method, apparatus. One example method includes: a first data analytics network element receive a first request from a sec data analytics network element, wherein the first request carries an analytics type identifier and second model requirement information, and the first request requests information of a model that corresponds to the analytics type identifier and meets the second model requirement information. The second data analytics network element receives the information of the model from the first data analytics network element.

Abstract: This application provides a service traffic steering method and an apparatus, to improve user experience. The method includes: A first network device determines at least one second terminal device, where first quality of experience corresponding to a first service flow of the second terminal device does not meet target quality of experience corresponding to the first service flow of the second terminal device. The first network device determines a target radio access network device, where the target radio access network device meets the target quality of experience corresponding to the first service flow of the at least one second terminal device. The first network device connects the first service flow of the at least one second terminal device to the target radio access network device.

Abstract: A method for correlating data of a terminal device and an apparatus are provided, to correlate data of a single terminal device by a RAN side, and further analyze the data of the terminal device. The method is: A first device obtains first data of a terminal device on a second device, where the first data includes first correlation information, the first correlation information is used to correlate the first data with second data of the terminal device on a first access network device, the second data includes the first correlation information, and there are one or more terminal devices. The first device determines a first identifier of the terminal device, and sends the first correlation information and the first identifier of the terminal device to a second access network device.

Abstract: A method includes: a first network element receives first information used to indicate to instantiate a VNF. The first network element obtains hardware resource information used to indicate an available resource of a network functions virtualization infrastructure, where the available resource includes a resource of at least one piece of hardware, and the hardware represents a minimum physical resource set used to bear a virtual machine. The first network element determines, based on the first information and the hardware resource information, a one-to-one mapping relationship between at least one virtual machine corresponding to the VNF and at least one piece of hardware corresponding to the VNF. The first network element determines, based on the mapping relationship, to instantiate the at least one virtual machine on the at least one piece of hardware.