Regularization Techniques to Reduce Overfitting
A machine learning team is developing a model to predict server failure from telemetry data. They use a deep neural network with 3 hidden layers. After training, the model achieves 99% accuracy on training data but only 85% on validation data. Which technique should the team apply to reduce the generalization error?
⚠ Common exam trap
CompTIA often tests the distinction between techniques that address overfitting (regularization) versus those that address underfitting (more layers, higher learning rate) or data quantity, leading candidates to mistakenly choose adding more data or increasing model 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
✓
Apply L2 regularization
The model exhibits high variance (overfitting) because it achieves 99% accuracy on training data but only 85% on validation data. L2 regularization (also known as weight decay) adds a penalty proportional to the squared magnitude of the weights to the loss function, which discourages the network from fitting noise in the training data and improves generalization. This directly reduces the gap between training and validation performance.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Increase the number of hidden layers
Why it's wrong here
Adding hidden layers increases model capacity, worsening the train-validation gap rather than closing it. It is tempting because deeper networks can capture richer feature interactions, and would be the right move when both training and validation error are high, indicating underfitting rather than the overfitting seen here.
- ✓
Apply L2 regularization
Why this is correct
L2 regularization adds a penalty on large weights to the loss function, shrinking model complexity and curbing the overfitting behind the 99% training versus 85% validation gap. This directly reduces the generalization error the team needs to lower.
- ✗
Increase the learning rate
Why it's wrong here
Raising the learning rate worsens the train-validation gap by making optimisation overshoot, increasing rather than reducing generalization error. A larger rate suits escaping shallow local minima early in training; the 99% versus 85% split calls for regularisation such as dropout or weight decay.
- ✗
Add more training data
Why it's wrong here
The 14-point train-validation gap is variance, so additional data addresses it only marginally; the network already memorises training samples. More data is tempting because it genuinely helps when a model underfits or when the dataset is small, but here regularisation or dropout targets the overfitting directly.
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.