Courseiva

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

A data scientist trains a regression model to predict energy consumption for a smart building. The model achieves very low error on the training data but performs significantly worse on a held-out validation set. Which technique would most directly address this problem?

⚠ Common exam trap

Watch out — candidates often confuse cross-validation (a performance evaluation method) with a technique to fix overfitting, or think hyperparameter tuning alone resolves overfitting without understanding that regularization is the specific mechanism to penalize complexity.

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

Regularization

The model's low training error but high validation error indicates overfitting, where the model has memorized the training data rather than learning generalizable patterns. Regularization (e.g., L1 or L2) directly penalizes large coefficients, reducing model complexity and improving generalization to unseen data.

Answer analysis

Option-by-option breakdown

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

  • Feature engineering

    Why it's wrong here

    Feature engineering creates or transforms predictor variables to better capture underlying patterns in the data. While this can improve model accuracy, it does not constrain the model's hypothesis space or add a penalty for coefficient magnitude, so it does not directly counteract overfitting caused by excessively complex models. Overfitting from too many or overly flexible features would remain unless the feature set is aggressively simplified, which is not the primary mechanism of regularization.

  • Regularization

    Why this is correct

    Regularization directly addresses overfitting by adding a penalty term to the loss function that grows with the magnitude of the model's coefficients. Techniques such as L1 (Lasso) and L2 (Ridge) force the model to keep weights small or drive some to zero, effectively reducing model complexity and variance. This penalty discourages the model from fitting noise in the training data, which is exactly why it is the correct method for the scenario described.

  • Cross-validation

    Why it's wrong here

    Cross-validation is a resampling procedure used to estimate how well a model generalizes to unseen data by partitioning the dataset into training and validation folds. It does not change the model's learning algorithm or impose any constraint on the coefficients, so it cannot prevent overfitting itself. Rather, it helps detect overfitting by revealing a gap between training and validation performance, but the model complexity remains untouched and will still overfit if left unregularized.

  • Hyperparameter tuning

    Why it's wrong here

    Hyperparameter tuning systematically searches for the best configuration of settings like learning rate, tree depth, or number of layers. While it can indirectly reduce overfitting if a less complex hyperparameter setting is chosen, it does not specifically penalize coefficient size or add a structural constraint to the objective function. The correct, direct remedy for overfitting in a regression model is regularization, which modifies the loss function itself rather than merely selecting among pre-defined model configurations.

About these practice questions

One of 985 original AI-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →

How Courseiva writes practice questions · Editorial policy

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.