Method and system for elaborating model context by computing explanation-guided model similarity

A computing apparatus for obtaining contextual information about a machine learning model, the computing apparatus comprising:

• a processor;
• a memory;
• and
• a communication interface coupled to each of the processor and the memory,
• wherein the processor is configured to:
• receive, via the communication interface, at least a first set of raw data that is usable for training a first model;
• train the first model by using the at least the first set of raw data;
• compute a set of common background data based on the at least the first set of raw data;
• compute a first explanation based on an output of the first model and the computed set of common background data;
• compute an agnostic model representation of the first model;
• compute, based on the first explanation and the agnostic model representation, an explanation-driven representation of the first model;
• and
• determine, based on the explanation-driven representation, contextual information that relates to the first model.

A non-transitory computer readable storage medium storing instructions for obtaining contextual information about a machine learning model, the storage medium comprising executable code which, when executed by a processor, causes the processor to:

• receive at least a first set of raw data that is usable for training a first model;
• train the first model by using the at least the first set of raw data;
• compute a set of common background data based on the at least the first set of raw data;
• compute a first explanation based on an output of the first model and the computed set of common background data;
• compute, based on an output of the first model, an agnostic model representation of the first model;
• compute, based on the first explanation and the agnostic model representation, an explanation-driven representation of the first model;
• and
• determine, based on deep explanation-driven representation, contextual information that relates to the first model.