Different AI Models and Development Processes AIE Practice Questions (Page 2)

Questions 6–10

1. 6expert · hard

   A team has developed a model to predict customer churn. The model performs well on the validation set, but after deployment, the business reports that the model's predictions are not improving retention campaigns. The team suspects the model is not aligned with the business objective. What should they do first?

   • ACollect more data and retrain the model
   • BDeploy a more complex model
   • CRevisit the problem definition and success metrics with stakeholders
   • DIgnore the business feedback because the model is technically sound
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   In the AI development lifecycle, what is the purpose of iterative refinement?

   • ATo replace the model with a completely different type after every evaluation.
   • BTo deploy the model as quickly as possible without testing.
   • CTo repeatedly adjust the model based on evaluation results to improve performance.
   • DTo collect more data only once at the beginning.
   Reveal Solution Discussion Select an answer first
4. 8expert · hard

   A robotics company is developing a robot that must navigate a warehouse and avoid obstacles. The robot can sense its environment and receives a reward for reaching a destination without collisions. The team has no labeled data of optimal paths. Which model type and development approach is most appropriate?

   • AReinforcement learning with a reward function for reaching the destination and avoiding collisions
   • BUnsupervised learning to map the warehouse
   • CSupervised learning with a dataset of human-operated paths
   • DA pre-trained supervised model from a different domain
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   An e-commerce company wants to recommend products to users based on their browsing history. They have a large dataset of user-item interactions but no explicit ratings. The goal is to increase user engagement. Which model type and evaluation metric are most appropriate?

   • AReinforcement learning; evaluate using cumulative reward from user clicks
   • BCollaborative filtering (unsupervised); evaluate using click-through rate or engagement metrics
   • CUnsupervised clustering; evaluate using silhouette score on user groups
   • DSupervised regression; evaluate using mean squared error on predicted ratings
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   Why is model evaluation an important step in the AI development lifecycle?

   • AIt ensures the model performs well on unseen data and meets the desired performance criteria.
   • BIt is only needed for complex models, not simple ones.
   • CIt guarantees the model will be 100% accurate on all future data.
   • DIt eliminates the need for data preprocessing.
   Reveal Solution Discussion Select an answer first