How to minimize cost for model retraining?
Short answer
D Download the weather data each week, and download the flu data each month. Deploy the model to a Vertex AI endpoint with feature drift monitoring, and retrain the model if a monitoring alert is detected.
Question
Key Takeaways
The question tests cost optimization in MLOps, where the trap is selecting a fixed retraining schedule instead of a monitoring-based trigger.
To minimize retraining costs for a flu prediction model, use Vertex AI feature drift monitoring to trigger retraining only when necessary. The community consensus supports this dynamic approach over fixed schedules to optimize resource usage.
Common mistake: Choosing a fixed schedule (Options A, B, or C) because it is straightforward, failing to recognize that retraining only on drift saves significant costs.
Expert Analysis
Why the Answer Is Correct
Option D is the optimal solution because it aligns data ingestion with the publication frequency of the sources (weather weekly, flu monthly) and utilizes feature drift monitoring. By retraining only when a monitoring alert is detected, you ensure resources are spent only when the model's predictive power is likely compromised by data changes, directly minimizing costs.
Why the Other Options Are Wrong
Options A, B, and C propose fixed retraining schedules (weekly or monthly). While Option C aligns ingestion better than A, all fixed schedules risk incurring unnecessary compute costs by retraining even when the data distribution is stable and the model performance remains adequate. They lack the intelligence to adapt to actual data behavior.
Community Comment Notes
Community members emphasized that 'Selective Retraining' is the core concept here, noting that retraining should only occur when feature drift is detected. Comments also highlighted the importance of ingesting weather data weekly to capture potential trends, even if the target variable (flu stats) is only available monthly.
Official Reference
https://cloud.google.com/vertex-ai/docs/model-monitoring/overview
This link opens the official vendor documentation in a new tab.
Exam Strategy
For questions focused on minimizing operational costs in MLOps, prioritize solutions that use triggers (like monitoring alerts) over fixed time intervals. Ensure the data ingestion strategy matches the availability of the specific data sources.
Community Votes
100% of anonymous learners picked answer D. Votes are pick records left by other test-takers — they are not the verified answer.
Discussion 5 contributions
Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.
From public community threads 5
-
fitri001 👍 5 Answered DWeather Data Update: Downloading weather data weekly captures the latest trends potentially influencing flu infections. Flu Data Update: Downloading flu statistics monthly aligns with the data publication schedule and avoids unnecessary processing for data that might not have changed. Feature Drift Monitoring: Vertex AI endpoint monitoring helps identify significant changes in the weather data distribution (feature drift) over time. Retrain Based on Alerts: Retraining the model is triggered only when feature drift is detected, ensuring the model stays relevant without unnecessary retraining cycles.
-
b1a8fae 👍 4 Answered DD. This way, cost is minimized by only retraining when feature drift takes place.
-
pikachu007 👍 2 Answered DSelective Retraining: Retraining occurs only when necessary, triggered by feature drift alerts, reducing cloud resource usage and associated costs. Efficient Data Utilization: Weather data is downloaded weekly to capture potential changes, but model retraining waits for monthly flu data, ensuring model relevance without excessive updates. Early Drift Detection: Vertex AI's feature drift monitoring proactively identifies model performance degradation, prompting timely retraining to maintain accuracy.
-
guilhermebutzke 👍 1 Answered DMy Answer: D Even though the model predicts values for the next month, it is necessary to consume weekly data because the model's output could change based on new weekly data. Therefore, it is necessary to download data weekly and monthly. Furthermore, it is not necessary to retrain the model if the feature distribution remains unchanged.
-
pinimichele01 👍 1 Answered Dminimize cost
Reader comments & corrections
No reader corrections yet — spotted an error or have a note? Share it below.
Related Analysis
Practice All PMLE Questions
Access 65 questions with complete answers and detailed explanations.
View Full PMLE Practice Test →