Evolution CT-AI Practice Questions (Page 2)

Questions 6–10

1. 6expert · medium

   A team maintains a real-time recommendation system that is retrained weekly. After a recent retraining, the model's accuracy on the validation set improved, but the system's response time increased by 40%, and some users report that the recommendations are less relevant. The team must decide whether to roll back the update. Which combination of quality characteristics should the team evaluate to make the decision?

   • AFairness and robustness
   • BPerformance and reliability
   • CSecurity and privacy
   • DUsability and accessibility
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   How can the evolution of an AI-based system impact its reliability?

   • AEvolution has no effect on reliability because reliability is determined only by the initial design
   • BEvolution only affects performance, not reliability
   • CEvolution always improves reliability because the model is updated with more data
   • DEvolution can degrade reliability if the model's performance on certain inputs changes unexpectedly after retraining
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A team has deployed a predictive maintenance model for industrial equipment. Over time, the model's accuracy has declined because the equipment's operating conditions have changed. The team plans to retrain the model with new data. Which testing strategy should be used to ensure the retrained model maintains quality in the new operating conditions?

   • AOne-time regression testing
   • BContinuous validation with monitoring
   • CLoad testing
   • DAdversarial testing
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   Which testing strategy is specifically used to ensure that a newly updated AI model does not break existing functionality?

   • ARegression testing
   • BUsability testing
   • CLoad testing
   • DExploratory testing
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A team is planning to update a machine learning model that is used for medical diagnosis. The update is expected to improve sensitivity but may reduce specificity. The team wants to ensure that the update does not compromise patient safety. Which testing strategy should be prioritized?

   • ALoad testing to ensure the system can handle peak demand
   • BAdversarial testing to ensure robustness against malicious inputs
   • CRigorous regression testing on a comprehensive set of historical cases
   • DUsability testing to ensure clinicians can interpret the results
   Reveal Solution Discussion Select an answer first