Adapting automated assistant functionality based on generated proficiency measure(s)

A method implemented by one or more processors, the method comprising:

• generating a class proficiency measure that is specific to a class of interactions with an automated assistant and that is specific to one or both of:
• (i) a particular user, and (ii) a particular assistant device, wherein generating the class proficiency measure comprises:
• processing instances of data for past assistant interactions to generate individual metrics for the class, wherein processing the instances of data for the past assistant interactions is based on the past assistant interactions being of the class and being initiated by one or both of (i) the particular user and (ii) the particular assistant device, and
• generating the class proficiency measure as a function of the individual metrics;
• assigning, in one or more computer readable media, one or both of (i) the particular user and (ii) the particular assistant device in association with a selected subset of classes of interactions;
• subsequent to the assigning:
• processing audio data that is detected via one or more microphones of the particular assistant device and that captures a spoken utterance of the particular user;
• determining, based on the processing, that the spoken utterance is directed to the class of interactions;
• and
• in response to the spoken utterance being directed to the class of interactions, performing one or both of:
• providing a shortened response to the spoken utterance in lieu of a longer response, and
• bypassing providing a confirmation message before fulfilling a request embodied in the spoken utterance.

An assistant device, comprising:

• one or more microphones;
• one or more processors operable to execute instructions to:
• generate, for each of a plurality of disparate classes of interactions with an automated assistant, an associated class proficiency measure that is specific to the assistant device, wherein in generating each of the associated class proficiency measures one or more of the processors are to:
• process corresponding instances of data for corresponding past assistant interactions, based on the corresponding past assistant interactions being for a corresponding one of the disparate classes of interactions and being initiated by the assistant device;
• select a first set of one or more of the disparate classes of interactions for inclusion in a subset of classes of interactions based on their associated class proficiency measures satisfying a threshold;
• in response to selecting the first set:
• purge, from on-device storage of the particular assistant device, one or more machine learning models or grammars that are specific to any of the disparate classes of interactions not included in the first set.