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 science team trains several machine learning models for a regression task. They observe that Model A has low training error and low test error. Model B has low training error but high test error. Model C has high training error and high test error. Which model would most likely benefit from an ensemble technique that averages the predictions of multiple models?

⚠ Common exam trap

A common mix-up: candidates assume ensembles always improve accuracy, but they are most effective for high-variance (overfit) models, not for underfit or already well-generalized models.

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

✓

Model B (low training error, high test error)

Model B exhibits low training error but high test error, which is a classic sign of overfitting. Ensemble techniques like averaging predictions from multiple models reduce variance and improve generalization, making them most beneficial for overfit models. In Azure Machine Learning, you can use an ensemble pipeline or AutoML's VotingEnsemble to combine diverse models and lower test error.

Answer analysis

Option-by-option breakdown

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

✗
Model A (low training error, low test error)
Why it's wrong here
Model A already exhibits strong generalization: both training and test error are low and comparable, indicating that the learned patterns transfer well to unseen data. While an ensemble could theoretically reduce variance further, the model's variance is already modest, so the marginal improvement would be small relative to the added complexity and compute cost. The more urgent priority would be feature engineering or calibration, not averaging this particular model.
✓
Model B (low training error, high test error)
Why this is correct
Model B has low training error but high test error, a classic sign of overfitting caused by high variance: the model memorizes noise in the training set and fails to generalize. Ensemble techniques such as bagging train multiple models on different bootstrap samples and average their predictions, which cancels out independent errors and stabilizes the decision boundary. This variance reduction directly targets the gap between training and test error, making Model B the clearest candidate for improvement.
✗
Model C (high training error, high test error)
Why it's wrong here
Model C shows high error on both training and test sets, which indicates underfitting due to high bias: the model is too simple to capture the underlying patterns. Averaging multiple such underfit models in an ensemble would still leave each individual with low capacity, so the aggregate remains biased (unless using a boosting strategy specifically designed to reduce bias). The appropriate fix is to increase model capacity, add relevant features, or reduce regularization, not to apply a generic averaging ensemble.
✗
None of the models would benefit from an ensemble technique
Why it's wrong here
This choice is incorrect because Model B's high-variance overfitting pattern is precisely the situation where ensemble methods provide the most benefit. Averaging or voting across multiple diverse models reduces the variance component of error, which is the dominant problem for Model B. Since at least one model clearly benefits, the statement that none would benefit is false.

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