QuestionQ29

VerifiedML Model Development

A company is developing a deep learning model on Amazon SageMaker. The company uses a large volume of data for the training dataset. The company must optimize the model’s hyperparameters to minimize the loss function on the validation dataset.

Which hyperparameter tuning strategy will achieve this goal with the LEAST computation time?

Explanation

Hyperband uses intermediate training results to stop underperforming hyperparameter configurations early and dynamically allocate resources to promising configurations. For large training jobs, this reduces computation time while optimizing the selected validation-loss objective.

Learn more

Community Discussion

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