Hyperparameter tuning is the practice of running training experiments with different hyperparameter settings to find values that improve model performance. The process may optimize accuracy, loss, robustness, efficiency, or another objective. It differs from learning model parameters because hyperparameters are set outside the training data-fitting process.
Hyperparameter tuning in Vertex AI involves running multiple trials of a training application with different values for the chosen hyperparameters, set within specified limits. The goal is to optimize the hyperparameter settings to maximize the model's predictive accuracy. For more information, see Hyperparameter tuning overview.