Describe fundamental principles of machine learning on Azure →hardMultiple ChoiceObjective-mapped

AI-900 Practice Question: Describe fundamental principles of machine learning on Azure

A data scientist evaluates a regression model that predicts house prices. On the test set, the Mean Absolute Error (MAE) is $8,000 and the Root Mean Squared Error (RMSE) is $25,000. What does the large difference between MAE and RMSE indicate about the model's errors?

⚠ Common exam trap

Many candidates confuse the mathematical behavior of RMSE (which amplifies large errors) with concepts like overfitting or bias, rather than recognizing it as a direct indicator of outlier errors in the predictions.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

The model has some predictions with very large errors

The large difference between MAE ($8,000) and RMSE ($25,000) indicates that the model has some predictions with very large errors. RMSE squares the errors before averaging, which heavily penalizes large deviations, so a significantly higher RMSE relative to MAE suggests the presence of outliers or extreme prediction errors in the test set.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
The model is overfitting the training data
Why it's wrong here
Overfitting is a diagnosis that requires comparing in-sample (training) performance with out-of-sample (test) performance. The reported metrics are exclusively test-set residuals, so they reveal nothing about the model's fit to the training data. Even a high RMSE relative to MAE on the test set can occur without any overfitting, as it simply indicates the distribution of test errors is skewed by large residuals. Therefore, these metrics alone cannot support the overfitting claim.
✗
The model predictions are consistently biased high
Why it's wrong here
Bias refers to the sign and magnitude of the mean prediction error (e.g., mean(y_pred - y_true)), whereas both MAE and RMSE are computed from absolute or squared errors, discarding the direction of the mistakes. Two models with identical MAE and RMSE can have entirely different bias: one might consistently overpredict while another underpredicts. Consequently, the gap between RMSE and MAE contains no information about whether predictions systematically overshoot or undershoot the target.
✓
The model has some predictions with very large errors
Why this is correct
Because RMSE squares each error before averaging and taking the square root, it weights large residuals disproportionately, whereas MAE treats all errors linearly. When RMSE substantially exceeds MAE, the error distribution must have a heavy positive tail: most errors are moderate, but a few predictions deviate very far from the true values. This is a direct, test-set-based inference about error magnitude, not about training dynamics or systematic bias. In practice, a ratio RMSE/MAE much greater than 1 signals the presence of influential outliers among the predictions.
✗
The model has high variance due to outliers in training data
Why it's wrong here
High variance (overfitting) is a possible cause, but the given test set metrics alone cannot differentiate between high variance and other causes. The difference in MAE and RMSE directly points to large errors in prediction, not specifically to the source of those errors.

Go deeper

AI-900 Practice Question: Describe fundamental principles of machine learning on Azure

Why each option matters

Option-by-option breakdown

Related to this question