You are a machine learning specialist working for a credit card company where you are building a fraud detection model. You have your model built using the XGBoost algorithm, and you are now attempting to find the best version of your model by performing automatic model tuning. You are creating your hyperparameter tuning job, and you need to select the appropriate technique the job will use to find your best hyperparameters. You wish to run the least number of hyperparameter tuning training jobs as possible.
Which hyperparameter tuning technique is best suited to your requirements?

Answer options:

A.Bayesian optimization
B.Bayesian classification
C.Random search
D.Logistic regression

Answer: A
Option A is correct. The Bayesian optimization approach to hyperparameter tuning results in fewer tuning job runs than the random search method.
Option B is incorrect. The Bayesian technique used by the hyperparameter tuning job in SageMaker is Bayesian optimization, which solves the problem using regression, not classification.
Option C is incorrect. A random search method is a valid option when using SageMaker hyperparameter tuning. But the random search approach usually requires running many more training jobs to get the best hyperparameters when compared to the Bayesian optimization approach. You have a requirement to run the least number of hyperparameter tuning training jobs as possible.
Option D is incorrect. Logistic regression is not a valid option when using SageMaker hyperparameter tuning.
Reference:
Please see the Amazon SageMaker developer guide titled How Hyperparameter Tuning Works (https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-how-it-works.html), the AWS Machine Learning blog titled Amazon SageMaker automatic model tuning now supports random search and hyperparameter scaling (https://aws.amazon.com/blogs/machine-learning/amazon-sagemaker-automatic-model-tuning-now-supports-random-search-and-hyperparameter-scaling/), and the AWS Machine Learning blog titled Simplify machine learning with XGBoost and Amazon SageMaker (https://aws.amazon.com/blogs/machine-learning/simplify-machine-learning-with-xgboost-and-amazon-sagemaker/)