AI0-001 AI Concepts and Foundations Practice Question
Exhibit
Refer to the exhibit. Model Training Log (Epoch 1-5): Epoch 1/5 - loss: 2.3026 - accuracy: 0.1000 - val_loss: 2.3026 - val_accuracy: 0.1000 Epoch 2/5 - loss: 2.3026 - accuracy: 0.1000 - val_loss: 2.3026 - val_accuracy: 0.1000 Epoch 3/5 - loss: 2.3026 - accuracy: 0.1000 - val_loss: 2.3026 - val_accuracy: 0.1000 Epoch 4/5 - loss: 2.3026 - accuracy: 0.1000 - val_loss: 2.3026 - val_accuracy: 0.1000 Epoch 5/5 - loss: 2.3026 - accuracy: 0.1000 - val_loss: 2.3026 - val_accuracy: 0.1000 Note: The dataset has 10 classes.
Refer to the exhibit. The model is a neural network for 10-class classification. The training log shows no improvement over 5 epochs. Which of the following is the most likely root cause?
⚠ Common exam trap
CompTIA often tests the distinction between symptoms of high learning rate (oscillation/divergence) and vanishing gradients (flat loss), so candidates mistakenly choose 'learning rate too high' when they see no improvement, but the key clue is the absence of oscillation or divergence in the loss curve.
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 suffering from vanishing gradients, preventing weight updates.
The training log shows no improvement over 5 epochs, which is a classic symptom of vanishing gradients in deep neural networks. When gradients become extremely small during backpropagation, weight updates are negligible, causing the loss to stagnate. This is especially common in deep networks with sigmoid or tanh activations, where gradients saturate in the tails of the activation function.
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 batch size is too large, making gradient updates insignificant.
Why it's wrong here
Large batch size may slow training but still show some decrease in loss.
- ✗
The output layer uses sigmoid activation instead of softmax.
Why it's wrong here
Wrong activation might affect output interpretation but accuracy would likely not be exactly 0.1.
- ✗
The learning rate is too high, causing the loss to oscillate.
Why it's wrong here
Typically high learning rate leads to divergence or oscillation, not flat loss.
- ✓
The model is suffering from vanishing gradients, preventing weight updates.
Why this is correct
Vanishing gradients can cause no learning, leading to constant loss and random accuracy.
About these practice questions
Courseiva writes every AI0-001 question from scratch — 754 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 →
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.