mediumMultiple Choice
AIF-C01 Practice Question: During model training, a data scientist notices…
During model training, a data scientist notices that the model performs very well on the training data but poorly on the test data. The scientist suspects high variance. Which technique is MOST likely to reduce the variance and improve test performance?
⚠ Common exam trap
AWS often tests the misconception that lowering the learning rate or reducing data fixes overfitting, but the correct approach is to reduce model complexity through regularization or feature selection.
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
✓
Apply regularization (e.g., L1 or L2)
High variance indicates the model is overfitting to the training data, capturing noise rather than the underlying pattern. Regularization (L1/L2) adds a penalty to the loss function for large coefficients, effectively constraining the model complexity and reducing variance, which improves generalization to unseen test 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.
- ✗
Lower the learning rate
Why it's wrong here
Lowering the learning rate changes optimisation step size, not model capacity; it can slow convergence and slightly smooth fitting but does not constrain the parameters that cause high variance. It is tempting because unstable training sometimes mimics overfitting, yet regularisation, dropout or more data address variance directly.
- ✗
Increase the number of features
Why it's wrong here
Adding features increases model capacity, giving the learner more ways to fit training noise and typically worsening the train-test gap. It is tempting because extra features help when the model underfits, but here the model already fits training data well, so the requirement is to reduce variance, not raise capacity.
- ✓
Apply regularization (e.g., L1 or L2)
Why this is correct
High variance means the model memorises training noise, so it generalises poorly. L1 or L2 regularization penalises large weights, constraining model complexity and reducing that variance, which directly addresses the gap between strong training and weak test performance described in the stem.
- ✗
Decrease the amount of training data
Why it's wrong here
Reducing training data removes the very signal that constrains the hypothesis space, so the model overfits the smaller sample even harder and test performance drops. It is tempting because less data speeds up training, but variance reduction requires more representative data or explicit regularisation, not fewer examples.
Go deeper
Related to this question
About these practice questions
One of 862 original AIF-C01 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 →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AIF-C01 practice question is part of Courseiva's free Amazon Web Services 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 AIF-C01 exam.