Courseiva

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

What is 'ensemble learning' in machine learning and why does it improve performance?

⚠ Common exam trap

It's easy for candidates to confuse ensemble learning with model selection (C) or stability testing (D), as candidates often think picking the 'best' model or running multiple trials is the same as combining 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

Combining predictions from multiple models to improve accuracy and robustness

Ensemble learning combines predictions from multiple models (e.g., bagging, boosting, stacking) to reduce variance, bias, or improve robustness. By aggregating diverse models, it often achieves higher accuracy than any single model, as errors from individual models are averaged out or corrected. This is a core technique in Azure Machine Learning, where ensembles like Random Forest or Gradient Boosting are commonly used.

Answer analysis

Option-by-option breakdown

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

  • Combining predictions from multiple models to improve accuracy and robustness

    Why this is correct

    Ensemble learning explicitly aggregates the predictions of multiple diverse base models through techniques such as bagging, boosting, or stacking. Bagging (e.g., random forests) trains models in parallel and averages/votes to reduce variance, boosting (e.g., gradient boosting) trains sequentially to reduce bias, and stacking uses a meta-learner to combine model outputs. This diversity of models causes individual errors to cancel out, yielding better accuracy and robustness than any single model, which is the defining characteristic of an ensemble.

  • Training a single very large model on the full dataset without any data splitting

    Why it's wrong here

    Training one very large model on the full dataset without data splitting describes a monolithic supervised-learning workflow, not ensemble learning. Using all data for training maximizes the available signal for that one model but produces a single prediction function, with no diversity of models whose outputs could be combined. Data splitting is a separate matter of evaluating generalization; regardless of how data is split, the absence of multiple models and a combination mechanism means this option does not match the definition of an ensemble.

  • Selecting the best model from a group of candidates after evaluation

    Why it's wrong here

    Selecting the best model from a group of candidates after evaluation describes standard model selection or hyperparameter tuning, where you compare candidates on a validation set and keep only the winner. Ensemble learning, by contrast, deliberately retains all candidate models (or a diverse subset) and aggregates their predictions, often achieving performance that exceeds the best individual candidate. Choosing a single model discards potentially useful complementary information, which is precisely the opposite of ensembling, where the combination is the source of strength.

  • Running the same model multiple times with different random seeds to test stability

    Why it's wrong here

    Running the same model multiple times with different random seeds is a stability or reproducibility check, used to measure how sensitive a model is to random initialization or data shuffling. Here the goal is to assess variance in performance, not to improve predictions, and the runs are typically inspected separately rather than combined into a single prediction. While one could ensemble the resulting models, doing so would be a weak form of ensembling because all runs share the same architecture and data, lacking the model diversity that makes ensembles effective.

About these practice questions

Courseiva writes every AI-900 question from scratch — 985 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. 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.