AI-900 Practice Question: Describe fundamental principles of machine learning on Azure
A data science team trains a regression model to predict house prices. They evaluate the model using Mean Absolute Error (MAE). After deployment, they notice that the model occasionally produces large errors (e.g., underpredicting a luxury home by $500,000) while most predictions are within $20,000. The business is more concerned about the impact of these large errors than the average small error. Which additional metric should the team use to better capture the penalty for large errors?
⚠ Common exam trap
A common mix-up: candidates choose MAE or MAPE because they seem intuitive for 'average error,' but they fail to recognize that RMSE's squared term is specifically designed to penalize large outliers, which is the exact business concern described.
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
✓
Root Mean Squared Error (RMSE)
Root Mean Squared Error (RMSE) is the correct additional metric because it squares the residuals before averaging, which heavily penalizes large errors like the $500,000 underprediction. Unlike MAE, which treats all errors equally, RMSE amplifies the impact of outliers, making it a better fit for a business that cares more about catastrophic failures than typical small errors. This aligns with the need to capture the penalty for large deviations in regression model evaluation.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Root Mean Squared Error (RMSE)
Why this is correct
Root Mean Squared Error (RMSE) is the correct choice for penalizing large errors because it squares each residual before averaging and then takes the square root. Squaring amplifies the contribution of large errors: for example, a $50,000 error contributes 25 times more to the squared error sum than a $10,000 error, not just 5 times. This matches the requirement that large dollar errors in a house price prediction be treated as disproportionately damaging.
- ✗
R-squared
Why it's wrong here
R-squared measures the proportion of the total variance in the target that the model explains, not the typical magnitude of prediction errors in the original unit (dollars). Because it is a relative, scale-free ratio of residual sum of squares to total sum of squares, a model can have a high R-squared yet still produce large absolute RMSE values if the overall variance in house prices is high. It also doesn't specifically amplify large errors more than small errors; it summarizes overall fit rather than penalizing outlier residuals.
- ✗
Mean Absolute Percentage Error (MAPE)
Why it's wrong here
Mean Absolute Percentage Error (MAPE) expresses each error as a percentage of the actual house price, so the same $50,000 absolute error becomes a much smaller percentage when the actual price is $1,000,000 than when it is $200,000. This means MAPE does not consistently penalize large dollar errors and can make a costly misprediction on an expensive home look deceptively acceptable. It also treats all percentage errors equally regardless of the underlying dollar magnitude, and it is undefined or unstable when actual values are zero or near zero.
- ✗
F1 score
Why it's wrong here
F1 score is a classification metric computed from precision and recall, which in turn rely on true positives, false positives, and false negatives from a confusion matrix. A regression model predicting a continuous house price has no discrete class labels or decision threshold, so those counts—and therefore F1—cannot be meaningfully calculated. It measures the balance between false positives and false negatives for categorical outcomes, not the distance between predicted and actual numeric values.
Go deeper
Related to this question
Learn chapter
Regression and Classification
Key term
Model
In IT and AI, a model is a trained mathematical representation that learns patterns from data to make predictions or decisions.
Key term
Regression
Regression is a type of machine learning algorithm that predicts a continuous numeric output based on input data, used to model relationships between variables.
About these practice questions
This AI-900 question is part of Courseiva's 985-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.