AutoML and ML Runtime MACHINE-LEARNING-ASSOCIATE Practice Questions (Page 1)

Questions 1–5

1. 1application · medium

   A business unit wants to quickly prototype a model to predict inventory demand. They have a clean dataset and a tight deadline. Which approach best reduces the time from data to a working model?

   • AWrite custom Python code to train a model from scratch.
   • BUse Databricks AutoML to automatically generate and evaluate models.
   • CWait for the next scheduled release of a new ML library.
   • DUse a pre-trained model from a model zoo and fine-tune it.
   Reveal Solution Discussion Select an answer first
2. 2foundation · easy

   A data scientist provides a dataset with 50 columns to Databricks AutoML. How does AutoML typically handle the feature set during model training?

   • AIt uses all 50 features without any selection
   • BIt requires the user to manually specify the top 10 features
   • CIt automatically evaluates and selects the most relevant features
   • DIt randomly selects 10 features for each trial
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   What is the main purpose of AutoML's automated model selection process?

   • ATo automatically deploy the selected model to production
   • BTo generate a single model without any configuration options
   • CTo explore a variety of algorithms and hyperparameters to find the best-performing model
   • DTo replace the need for any human oversight in model development
   Reveal Solution Discussion Select an answer first
5. 4expert · hard

   A company wants to democratize ML across its organization. They have a mix of data scientists and business analysts. They want to ensure that business analysts can build models while data scientists can still have advanced control. What is the best strategy?

   • ARequire all users to use AutoML exclusively.
   • BProvide AutoML to business analysts and allow data scientists to use AutoML notebooks as a starting point for custom development.
   • CRestrict AutoML to data scientists only.
   • DProvide separate tools for analysts and data scientists.
   Reveal Solution Discussion Select an answer first
6. 5expert · hard

   A team is running AutoML on a cluster that uses the standard Databricks Runtime. They are experiencing performance issues and suspect that the runtime is not optimized for ML workloads. What should they do?

   • AReduce the dataset size to speed up AutoML.
   • BUse a different AutoML tool that is more efficient.
   • CSwitch the cluster to use the Databricks ML Runtime.
   • DIncrease the cluster size to compensate for the lack of optimization.
   Reveal Solution Discussion Select an answer first