mediumMultiple Choice
PDE Practice Question: A retail company uses a machine learning model to…
A retail company uses a machine learning model to predict inventory demand. The model is retrained weekly using Vertex AI Pipelines. Recently, the model's accuracy has degraded because the data distribution has shifted. Which action should you take to monitor and detect this drift automatically?
⚠ Common exam trap
Google Cloud often tests the distinction between monitoring model performance metrics (like MAE) versus monitoring input data distributions (feature drift), and candidates mistakenly choose a performance-based alerting option because they think accuracy degradation is the only signal, ignoring that drift detection is the proactive mechanism to catch the root cause before accuracy drops.
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 for the endpoint and configure alerting on feature drift
Vertex AI Model Monitoring is purpose-built to automatically detect feature drift and prediction drift on deployed endpoints. By enabling it and configuring alerting on feature drift, you can proactively identify when the distribution of incoming features deviates from the training data, which directly addresses the root cause of accuracy degradation without manual intervention.
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 for the endpoint and configure alerting on feature drift
Why this is correct
Vertex AI Model Monitoring on the endpoint computes feature distribution statistics against a training baseline and raises alerts when skew or drift exceeds configured thresholds. This detects the shifted data distribution automatically, satisfying the requirement to monitor and detect drift without manual inspection.
- ✗
Set up alerts for when the model's mean absolute error exceeds a threshold on the evaluation dataset
Why it's wrong here
Mean absolute error on the evaluation dataset measures historical model performance, not shifts in incoming feature distributions, so drift goes undetected until labels arrive. It is tempting because it monitors accuracy, and would be correct for tracking overall model quality degradation over time.
- ✗
Enable Cloud Logging for the prediction endpoint and search for error logs
Why it's wrong here
Cloud Logging captures endpoint errors and request metadata, not statistical changes in feature values, so distribution shift remains invisible. It is tempting because logging is easy to enable, and would be correct for diagnosing prediction failures or latency issues rather than data drift.
- ✗
Schedule a job to compare the distribution of incoming features with the training data using Cloud Dataflow
Why it's wrong here
Cloud Dataflow is a batch and stream processing service; it does not compute feature distribution comparisons or emit drift metrics automatically. It is tempting as a data pipeline tool, and would be correct for transforming or moving data, not for detecting training-serving skew.
Go deeper
Related to this question
About these practice questions
One of 747 original PDE 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 PDE 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 PDE exam.