Answer-first summary for fast verification
Answer: Median stopping, Truncation selection
The question requires two early termination policies that: (1) account for all previous runs when evaluating the current run, and (2) avoid comparing only with the best performing run to date. Median stopping (A) computes running averages across all training runs and stops runs worse than the median, satisfying both conditions. Truncation selection (D) cancels a percentage of lowest performing runs at each evaluation interval by comparing against all runs, also meeting both requirements. Bandit policy (B) compares against the best run plus a slack factor, violating condition 2. Default policy (C) means no early termination, violating both conditions. The community consensus strongly supports AD, with the highest-voted comment (25 upvotes) providing clear reasoning that A and D satisfy both conditions while B and C do not.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You are training machine learning models in Azure Machine Learning and using Hyperdrive for hyperparameter tuning. In previous runs, many models have shown similar performance.
You need to select an early termination policy that meets the following requirements:
Which two early termination policies should you use? Each correct answer presents part of the solution. NOTE: Each correct selection is worth one point.
A
Median stopping
B
Bandit
C
Default
D
Truncation selection
No comments yet.