Machine learning model validation with verification data

A method of wireless communication by a first network device, comprising:

• transmitting a network configuration to a second network device in response to a change in user equipment (UE) environment;
• transmitting machine learning model verification data to the second network device in accordance with the network configuration;
• receiving a report, from the second network device, indicating performance of a machine learning model with the verification data;
• validating the machine learning model in response to the performance satisfying a performance condition;
• transmitting first instructions, to the second network device, to activate the machine learning model in response to the performance level exceeding a threshold level;
• and
• transmitting second instructions, to the second network device, to fall back to a non-machine learning model in response to the performance level being below the threshold level.

An apparatus for wireless communication by a user equipment (UE), comprising:

• at least one memory;
• at least one wireless transceiver;
• and
• at least one processor coupled to the at least one memory and the at least one wireless transceiver, the at least one processor configured:
• to wirelessly receive a network configuration from a network device in response to a change in UE environment;
• to wirelessly receive a machine learning model at the UE in accordance with the network configuration, the machine learning model being unvalidated and trained;
• to quantize inputs to the machine learning model;
• to compile the machine learning model;
• to wirelessly obtain verification data from the network device;
• to validate the machine learning model with the verification data to determine a performance level of the machine learning model;
• to wirelessly receive first instructions, from the network device, to activate the machine learning model in response to the performance level exceeding a threshold level;
• to wirelessly receive second instructions, from the network device, to fall back to a non-machine learning model in response to the performance level being below the threshold level;
• and
• to communicate with the network device in accordance with inferences from the machine learning model in response to the performance level exceeding the threshold level or in accordance with processing by the non-machine learning model in response to the performance level being below the threshold level.

An apparatus for wireless communication by a first network device, comprising:

• at least one memory;
• at least one wireless transceiver;
• and
• at least one processor coupled to the at least one memory and the at least one wireless transceiver, the at least one processor configured:
• to transmit a network configuration to a second network device in response to a change in user equipment (UE) environment;
• to transmit machine learning model verification data to the second network device in accordance with the network configuration;
• to receive a report, from the second network device, indicating performance of a machine learning model with the verification data;
• to validate the machine learning model in response to the performance satisfying a performance condition;
• to transmit first instructions, to the second network device, to activate the machine learning model in response to the performance level exceeding a threshold level;
• and
• to transmit second instructions, to the second network device, to fall back to a non-machine learning model in response to the performance level being below the threshold level.