easyMultiple Choice
PMLE Prediction drift Practice Question
A company has deployed a fraud detection model on Vertex AI Prediction. After three months, the model's accuracy has degraded, and the business is losing money due to undetected fraud. What should the team implement to proactively detect such issues?
⚠ Common exam trap
A common trap is to confuse monitoring with logging or retraining. Logging provides data but not automated drift detection; retraining fixes drift but doesn't detect it proactively.
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
✓
Enable Vertex AI Model Monitoring to track prediction drift and alert when metrics exceed thresholds.
Vertex AI Model Monitoring tracks prediction drift and alerts when metrics exceed thresholds, enabling proactive detection of model degradation. Option B is wrong because Cloud Logging captures requests and responses but does not automatically detect drift, requiring manual review. Option C is wrong because shuffling training data does not help detect drift. Option D is wrong because scheduling retraining without monitoring cannot proactively detect issues before they cause loss.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Enable Vertex AI Model Monitoring to track prediction drift and alert when metrics exceed thresholds.
Why this is correct
Vertex AI Model Monitoring continuously compares incoming prediction requests against the training baseline, detecting feature skew and prediction drift. This satisfies the stem's requirement to proactively detect degradation before financial loss accumulates, triggering alerts when drift metrics breach configured thresholds so the team can retrain the fraud model.
- ✗
Set up Cloud Logging to capture all prediction requests and responses for manual review.
Why it's wrong here
Cloud Logging captures raw request and response payloads for retrospective inspection; it computes no drift, skew, or accuracy statistic, so degradation stays invisible until someone reads logs. It is tempting as a general observability tool, and would suit audit trails or debugging individual predictions rather than proactive model monitoring.
- ✗
Randomly shuffle the training data before retraining to improve robustness.
Why it's wrong here
Shuffling training data addresses neither input drift nor concept drift, so it cannot surface the accuracy decay occurring in production. It is tempting because shuffling removes ordering bias during training, which is the right step when datasets arrive sorted by label or time — but that is a preprocessing concern, not ongoing monitoring.
- ✗
Schedule a monthly job to retrain the model with the latest data without monitoring.
Why it's wrong here
Retraining monthly without monitoring cannot detect degradation between runs, so fraud losses continue unnoticed until the next cycle. It is tempting because scheduled retraining addresses data drift, and would suit a scenario where drift is already measured and retraining cadence is the only remaining task.
Go deeper
Related to this question
About these practice questions
One of 775 original PMLE practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PMLE practice question is part of Courseiva's free Google Cloud 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 PMLE exam.