Question 40
Domain 4: Guidelines for Responsible AIA company has an ML model. The company wants to know how the model makes predictions. Which term refers to understanding model predictions?
Correct answer: A
Explanation
Model interpretability means the model’s decision logic can be understood and audited. The source says traditional ML models are “fully interpretable” and that “feature importance and decision logic can be extracted and audited,” which matches understanding how predictions are made.
Why each option is right or wrong
A. Model interpretability
The question asks for the term used when a team wants to understand how a model arrives at its outputs; in the source, traditional ML models are described as “fully interpretable” because their feature importance and decision logic can be extracted and audited. That maps directly to the concept of interpretability, not just general performance or accuracy, and it is the property that lets you inspect the prediction path for a specific model decision.
B. Model training
Model training is the process of fitting a model to data, not explaining its predictions.
C. Model interoperability
Model interoperability is the ability to work across systems or models, not to explain outputs.
D. Model performance
Model performance measures accuracy or error, not whether predictions are understandable.