AI Concepts CERTIFIED-TECHNOLOGY-SOLUTIONS-PROFESSIONAL Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   Which stage of the AI data lifecycle involves converting raw data into a clean, usable format for model training?

   • AData labeling
   • BData preprocessing
   • CData ingestion
   • DData versioning
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A manufacturing company has deployed a predictive maintenance model that monitors equipment sensors. After a few months, the model's accuracy begins to decline because the equipment behavior changes over time. What should the company do to maintain model performance?

   • AReplace the model with a rule-based system
   • BIncrease the model's complexity to fit old data better
   • CStop monitoring the model to avoid false alarms
   • DRetrain the model periodically with new data
   Reveal Solution Discussion Select an answer first
4. 3application · medium

   A self-driving car company is collecting sensor data from test vehicles. The data is used to train perception models. The company needs to ensure that the training data remains consistent and reproducible across different experiments. Which practice is essential?

   • ASharing data with external partners without restrictions
   • BStoring all data in a single large file for simplicity
   • CDeleting old data to save storage space
   • DVersioning the dataset and tracking changes over time
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A healthcare startup is building an image-classification model to detect anomalies in X-rays. They have collected a large dataset of images, but the data is unstructured and contains many duplicates and mislabeled files. The team is about to start training. What should they do first to improve the model's accuracy and avoid wasted compute?

   • ADeploy the model immediately to production to gather real-world feedback
   • BPerform data preprocessing and labeling to clean and annotate the dataset
   • CBegin training immediately to establish a baseline model
   • DIncrease the number of GPUs to speed up training
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A company is training a deep learning model on a multi-GPU cluster. The training process is long, and the company wants to minimize the risk of losing progress if a GPU fails or the cluster is interrupted. Which technique should they implement?

   • AData caching to speed up data loading
   • BRegular checkpointing of model weights and optimizer state
   • CIncreasing the number of GPUs to reduce training time
   • DGPU direct storage to bypass the CPU
   Reveal Solution Discussion Select an answer first