What feature of Amazon SageMaker allows for automatic model tuning?

The feature of Amazon SageMaker that enables automatic model tuning is hyperparameter optimization. This process is essential in machine learning because the performance of a model can be significantly impacted by the choice of hyperparameters, which are parameters that are not learned during training but instead set before the training process begins.

Hyperparameter optimization in SageMaker automates the search for the best hyperparameter configurations by running multiple training jobs with different combinations of hyperparameters. It employs techniques such as Bayesian optimization to efficiently find configurations that lead to improved model performance. This means that users can achieve better results without the need for extensive manual tuning, saving time and resources while improving model accuracy and efficiency.

In contrast, the other options refer to different aspects of the machine learning process. Data normalization involves preparing data to ensure that features are on a similar scale, which helps improve model performance but does not inherently involve automatic tuning. Feature engineering is the process of selecting and transforming variables into features that can be used in modeling, which is also a critical step but separate from hyperparameter optimization. Model validation focuses on evaluating the performance of the model after training, ensuring it generalizes well to unseen data, and does not relate to the tuning of hyperparameters during the training phase.