AI0-001 Machine Learning and Deep Learning Practice Question
A machine learning engineer notices that a linear regression model has high bias. Which action is most likely to reduce bias?
⚠ Common exam trap
CompTIA often tests the bias-variance tradeoff by making candidates confuse bias-reduction techniques with variance-reduction techniques, such as regularization or reducing training data, which actually increase bias or do not affect it.
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
✓
Use a more complex model, such as polynomial regression
High bias indicates that the model is too simple to capture the underlying patterns in the data, leading to underfitting. Using a more complex model, such as polynomial regression, increases the model's capacity to fit the training data better, directly addressing the underfitting issue. This is the standard approach to reduce bias in machine learning.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Use a more complex model, such as polynomial regression
Why this is correct
High bias means the model underfits, so increasing capacity with polynomial regression lets it capture non-linear relationships the linear hypothesis cannot represent, reducing bias. Note this typically raises variance, trading one error source for the other.
- ✗
Reduce the number of training samples
Why it's wrong here
Fewer training samples shrink the data available to fit the model, typically increasing bias rather than reducing it. It is tempting when addressing variance, where more data helps and less data hurts; high bias instead calls for a more expressive model or additional informative features.
- ✗
Add L2 regularization
Why it's wrong here
L2 regularization penalises large weights, shrinking model complexity and therefore increasing bias. It is tempting because regularization is a standard remedy for overfitting, but that is a variance problem; high bias needs a more flexible model or better features, not added constraint.
- ✗
Apply feature scaling
Why it's wrong here
Feature scaling changes feature magnitudes, not the model's functional form, so it cannot reduce bias from an overly simple hypothesis. It is tempting because scaling aids gradient descent convergence and distance-based algorithms, but for high bias the fix is a richer model or added relevant features.
About these practice questions
This AI0-001 question is part of Courseiva's 962-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI0-001 practice question is part of Courseiva's free CompTIA 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 AI0-001 exam.