DP-100 practice question 24 of 35

You train a classification model by using a logistic regression algorithm. You must be able to explain the model’s predictions by calculating the…

Domain Implement responsible machine learning

Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while others might not have a correct solution. After you answer a question in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen. You train a classification model by using a logistic regression algorithm. You must be able to explain the model’s predictions by calculating the importance of each feature, both as an overall global relative importance value and as a measure of local importance for a specific set of predictions. You need to create an explainer that you can use to retrieve the required global and local feature importance values. Solution: Create a MimicExplainer. Does the solution meet the goal?

Answer options

A: Yes

B: No Correct answer

Correct Answer:

B. No

Explanation

The MimicExplainer is designed to explain complex, opaque "blackbox" models. It functions by training a simpler, intrinsically interpretable "surrogate" model (e.g., a decision tree) that learns to approximate the predictions of the original blackbox model. The explanations are then generated from this surrogate model. A logistic regression model is a "glassbox" model, meaning it is already intrinsically interpretable. Using a MimicExplainer is an incorrect and indirect approach, as it would provide explanations for the surrogate model, not the original logistic regression model itself. The appropriate tool for a linear model is the ShapLinearExplainer, which directly uses the model's structure to calculate feature importance accurately and efficiently.

Why Incorrect

A. Yes: This is incorrect. While a MimicExplainer can technically produce feature importance values, it is designed for blackbox models and would provide an indirect, approximated explanation for a simple glassbox model like logistic regression, making it an inappropriate solution.

References

1. Microsoft Azure Official Documentation: Model interpretability in Azure Machine Learning. Under the section "Interpretability with Azure Machine Learning

" the documentation provides a table of supported explainers.

MimicExplainer: "Explains black box models by training a global surrogate model that mimics the black box model." This confirms its purpose is for blackbox models.

ShapLinearExplainer: "Uses the SHAP framework to explain linear models." This is explicitly recommended for models like logistic regression. The text also states

"For linear models

we recommend ShapLinearExplainer." This directly contradicts the proposed solution.

2. Microsoft Azure Official Documentation: How to generate model explanations (SDK v1). In the "Create explainer object" section

the guidance is clear: "Use the ShapLinearExplainer to explain linear models." This reinforces that MimicExplainer is not the correct choice for the given scenario.

Keep practicing

Take the free DP-100 practice test

Ten exam-style questions with answers and explanations, plus the exam facts and study guides.

Start the free practice test