You manage an Azure Machine Learning workspace. You must provide explanations for the behavior of the models with feature importance measures. You need to configure a Responsible AI dashboard in…

The correct answer is D. Interpretability. The Interpretability component of the Responsible AI dashboard computes and visualizes feature importance scores - both global importance (which features most influence overall model behavior) and local importance (which features drove a specific prediction). Counterfactual…

Train and deploy models

Question

You manage an Azure Machine Learning workspace. You must provide explanations for the behavior of the models with feature importance measures. You need to configure a Responsible AI dashboard in Azure Machine Learning. Which dashboard component should you configure?

Options

ACounterfactual what-if
BCasual inference
CFairness assessment
DInterpretability

How the community answered

(35 responses)

A
3% (1)
C
3% (1)
D
94% (33)

Explanation

The Interpretability component of the Responsible AI dashboard computes and visualizes feature importance scores - both global importance (which features most influence overall model behavior) and local importance (which features drove a specific prediction). Counterfactual what-if shows the minimum input changes needed to flip a prediction. Causal inference estimates cause-and-effect relationships between features and outcomes. Fairness assessment evaluates performance disparities across demographic groups. Only Interpretability is specifically designed to explain model behavior through feature importance.

Topics

#Azure Machine Learning#Responsible AI#Model Interpretability#Feature Importance

Community Discussion

No community discussion yet for this question.

Full DP-100 Practice