ML UE capability and inability

Apparatus comprising:

• one or more processors, and
• memory storing instructions that, when executed by the one or more processors, cause the apparatus to:
• check whether a terminal indicates to a network its capability to execute and to train a machine learning model;
• monitor whether the terminal is in an inability state or an ability state;
• inform the network that the terminal is in the inability state if the terminal indicated the capability and the terminal is in the inability state, wherein, in the inability state, the terminal is not able to execute and train the machine learning model, or the terminal is not able to execute and train the machine learning model at least with a predefined performance;
• inform the network that the terminal is in the ability state if the terminal indicated the capability and the terminal is in the ability state, wherein, in the ability state, the terminal is able to execute and train the machine learning model at least with the predefined performance;
• update, by the terminal, a default behavior associated with the machine learning model when the inability state is detected, wherein the updating the default behavior comprises:
• updating, based on a memory full state, the default behavior to executing a non-machine learning algorithm, and
• updating, based on a memory state change from the memory full state, the default behavior to executing a light machine-learning algorithm requiring fewer measurements than the machine learning model;
• execute, by the terminal, the default behavior instead of the machine learning model if the terminal is in the inability state;
• supervise if the terminal executes and trains the machine learning model and goes into the inability state;
• request the network to instruct the terminal to stop executing and training the machine learning model if the terminal executes and trains the machine learning model and the terminal goes into the inability state;
• monitor if the terminal receives an instruction to execute and train the machine learning model;
• reject the instruction to execute and train the machine learning model if the terminal receives the instruction and the terminal is in the inability state;
• monitor whether the terminal goes from the inability state to the ability state;
• and
• inform the network that the terminal is in the ability state if the terminal indicated the capability and the terminal goes from the inability state to the ability state,
• wherein the terminal is in the inability state if all of the following conditions are fulfilled:
• a battery of the terminal is charged below a predefined threshold,
• a memory available for the executing and training of the machine learning model is less than a predefined memory threshold for the machine learning model, and
• a processing power for the executing and training of the machine learning model is less than a predefined processing power threshold for the machine learning model.