AI-102 Implement computer vision solutions Practice Question
You are using Azure AI Custom Vision to classify images of animals. The training set has 1000 images of cats and 1000 images of dogs. After training, the model performs well on the test set. However, when deployed, it misclassifies images of wolves as dogs. What is the most likely cause?
⚠ Common exam trap
Microsoft often tests the misconception that class imbalance is the primary cause of misclassification, but here the dataset is balanced, and the real issue is the lack of representative negative examples—a subtle but critical distinction in Custom Vision training.
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 training set does not include enough negative examples that look like dogs but are not.
The model misclassifies wolves as dogs because the training set lacks negative examples that are visually similar to dogs but belong to a different class. Custom Vision learns to distinguish classes based on the features present in the training images; without images of wolf-like canines labeled as 'not dog,' the model has no basis to reject wolves. This is a classic case of insufficient hard negative mining, where the model generalizes too broadly for the 'dog' class.
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 training set does not include enough negative examples that look like dogs but are not.
Why this is correct
The model learned dog features from images lacking wolf-like negatives, so it maps wolf visual traits onto the dog class. Adding negative examples resembling dogs but labelled otherwise would sharpen the decision boundary and satisfy the requirement to classify wolves correctly.
- ✗
The probability threshold is set too low.
Why it's wrong here
A low threshold changes confidence cut-offs, not the model's learned features; wolves are misclassified because training data contained no wolf images. It tempts because threshold tuning does affect precision-recall balance, and would be correct if valid classes were being rejected at inference.
- ✗
The model is overfitted to the training data.
Why it's wrong here
Overfitting would degrade test-set accuracy, yet the model performs well there; the failure is wolves being unseen classes. It tempts because overfitting genuinely harms generalisation, and would be correct if training and test accuracy diverged markedly.
- ✗
The training set has class imbalance.
Why it's wrong here
Both classes hold 1000 images, so no imbalance exists; the wolf misclassification stems from wolves being absent from training data. It tempts because imbalance genuinely degrades minority-class recall, and would be correct if one class had far fewer examples than the other.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AI-102 question from scratch — 761 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 AI-102 practice question is part of Courseiva's free Microsoft 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 AI-102 exam.