AI0-001 AI Concepts and Techniques Practice Question
A deep learning engineer is training a transformer model and notices that validation perplexity increases after a few epochs while training perplexity continues to decrease. Which of the following is the MOST likely cause?
⚠ Common exam trap
CompTIA AI often tests the distinction between optimization issues (like learning rate or batch size) and generalization issues (like overfitting), and the trap here is that candidates may confuse a rising validation loss with a learning rate that is too high, when in fact the divergence between training and validation metrics is the definitive clue for overfitting.
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
✓
The model is overfitting the training data
The described pattern—decreasing training perplexity alongside increasing validation perplexity—is the classic signature of overfitting. The model is memorizing the training data rather than learning generalizable patterns, causing its performance on unseen validation data to degrade after a certain point in training.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
The temperature parameter is set too high
Why it's wrong here
Temperature governs sampling during inference, not weight updates during training, so it cannot cause validation perplexity to rise across epochs. It tempts because high temperature does degrade output quality, making it the answer when a deployed model generates incoherent or overly random text.
- ✗
The batch size is too small
Why it's wrong here
A small batch size adds gradient noise, which typically destabilises training loss too; it does not produce the diverging train/validation pattern described. It tempts because small batches genuinely harm convergence speed and generalisation, and would be the answer if both losses were erratic.
- ✗
The learning rate is too low
Why it's wrong here
A low learning rate slows convergence; both training and validation perplexity would plateau rather than diverge. It tempts because under-training is a real problem, and it would be correct if the model had simply stopped improving on either dataset rather than overfitting the training set.
- ✓
The model is overfitting the training data
Why this is correct
Validation perplexity rising while training perplexity keeps falling is the classic divergence signature: the model memorises training-specific patterns rather than generalisable ones, so held-out performance degrades. That gap between the two curves is precisely what overfitting produces, making it the most likely cause here.
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.