You manage an Azure Machine Learning workspace. You configure a hyperparameter tuning run for an automated machine learning $($AutoML$)$ experiment.

The experiment must terminate if a new model does not perform better than the lowest performing trials.

You need to configure a policy that uses the percentage of lowest performing trials.

Which early termination policy should you select?

Select only one answer.

Bandit

Median stopping

Truncation selection

Evaluation interval