AI-300: Operationalizing Machine Learning and Generative AI SolutionsDemo

By Microsoft · Question Mode

QuestionQ23

Implement machine learning model lifecycle and operations

You train machine learning models in Azure Machine Learning and use Hyperdrive to tune hyperparameters.

In earlier model-training and tuning runs, many models delivered similar performance.

You need to choose an early-termination policy that meets these requirements:

  • accounts for the performance of all prior runs when assessing the current run
  • does not compare the current run only with the best-performing run so far

Which two early-termination policies should you use? Each correct answer presents part of the solution.

NOTE: Each correct selection is worth one point.

Choose two
  • A Bandit
  • B Default
  • C Median stopping
  • D Truncation selection
Explanation

Median stopping computes running averages across all training runs and terminates a run when its primary metric is worse than the median of those averages. Truncation selection compares runs by their primary metric and cancels a specified percentage of the lowest-performing runs at each evaluation interval. Both evaluate performance relative to the group of runs, whereas Bandit policy uses the most successful run as its reference.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!