AI0-001 AI Infrastructure and Technologies Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Lock the model version after initial deployment to ensure reproducibility and avoid unexpected behavior changes in production.

  Why it's wrong here

  Locking the model version prevents adaptation to drift, guaranteeing that accuracy will degrade as consumer behavior changes. Reproducibility is valuable for auditing, but it must be balanced with the need to update models when data distributions shift. This practice would leave the retail chain with increasingly inaccurate forecasts, undermining the purpose of the demand forecasting system.

• ✓

  Schedule regular retraining of the model on a rolling window of recent data, with validation against a holdout set before promotion to production.

  Why this is correct

  Periodic retraining on recent data allows the model to learn current demand patterns, seasonality, and promotion responses. A rolling window keeps the training set relevant while a holdout validation ensures the retrained model outperforms the incumbent before deployment. This directly counters concept drift and maintains forecast accuracy as consumer behavior evolves across the retail chain.

• ✗

  Increase the model's complexity by adding more layers and parameters to improve its ability to fit historical sales patterns.

  Why it's wrong here

  Adding complexity does not address drift; it may worsen it by overfitting to outdated patterns. A more complex model can memorize historical noise and fail to generalize when behavior shifts. The problem is not insufficient model capacity but changing data distributions. Without monitoring and retraining, added complexity increases maintenance cost and latency without solving the underlying accuracy degradation over time.

• ✓

  Set up automated data drift and concept drift monitoring with alerts when statistical properties of input features or prediction errors deviate from training baselines.

  Why this is correct

  Drift monitoring is essential because consumer behavior, weather patterns, and promotions change over time, causing the model's input distribution or the relationship between inputs and demand to shift. Automated alerts based on statistical tests like population stability index or KL divergence allow the team to detect degradation early and trigger retraining before forecast accuracy materially harms inventory decisions across stores.

• ✗

  Deploy the model as a batch job that runs once per quarter to reduce infrastructure costs and operational overhead.

  Why it's wrong here

  Quarterly batch runs are far too infrequent for daily demand forecasting that must react to weather and promotions. Forecasts would be stale for weeks, leading to stockouts or overstock. While batch processing can be cost-effective, the cadence must match business needs. This option reduces responsiveness rather than ensuring accuracy over time, and it does not address drift detection or adaptation.