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

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

A data scientist trains a regression model to predict house prices using features like bedrooms, square footage, and location. The model achieves a low error on the training data but performs significantly worse when used to predict prices in a new city with different property characteristics. Which concept best explains this poor performance?

⚠ Common exam trap

Candidates often confuse overfitting with the bias-variance tradeoff, but the question specifically asks for the concept that best explains the poor performance on new data, which is overfitting, not the general tradeoff.

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

✓

Overfitting

The model performs well on training data but poorly on new data from a different city, which is the classic symptom of overfitting. Overfitting occurs when a model learns noise and specific patterns in the training data that do not generalize to unseen data, especially when the new data has different characteristics (e.g., different property market dynamics). In this case, the model has memorized the training city's price patterns rather than learning generalizable relationships.

Answer analysis

Option-by-option breakdown

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

✗
Underfitting
Why it's wrong here
Underfitting describes a model too simple to capture the true structure of the training data, so it suffers from high bias and produces large errors on both the training set and unseen homes. In this scenario, the newly encountered homes would not perform worse because the model memorized the training set; rather, the model never learned the underlying price patterns in the first place.
✓
Overfitting
Why this is correct
Overfitting occurs when a model with excessive capacity learns not only the genuine patterns in the training data but also its random noise and idiosyncrasies. This results in very low training error but poor generalization, and the effect becomes especially visible when new houses come from a slightly different distribution, because the learned noise no longer matches. The symptom described—good performance during training but degraded performance on new data—is the classic signature of overfitting.
✗
Data leakage
Why it's wrong here
Data leakage happens when information from outside the training set, such as the target value or future data, is inadvertently used during training, inflating the model's apparent accuracy. Unlike the described behavior, leakage typically shows unrealistically strong performance during validation and even on some test sets, but not a sharp drop specifically on new homes unless the leakage source disappears. The question gives no indication of target information or temporal shortcuts leaking into the house-price features.
✗
Bias-variance tradeoff
Why it's wrong here
The bias-variance tradeoff is a general statistical framework that decomposes prediction error into bias, variance, and irreducible noise; overfitting is one manifestation of high variance, while underfitting reflects high bias. While the tradeoff describes the tension between model complexity and generalization, it is not itself a failure mode or the most direct diagnosis for a model that performs well on training data and poorly on new data. The behavior described is specifically overfitting, not the broader concept.

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