by Pier Paolo Ippolito – SAS Institute

Hyperparameter tuning is considered one of the most important steps in the machine learning pipeline and can turn, what may be viewed as, an “unsuccessful” model into a solid business solution by finding the right combination of input values. In this chapter, the theoretical foundations behind different traditional approaches to optimizing hyperparameters, such as Manual Search, Grid Search, and Random Search, will be laid out in addition to outlining more advanced and informed search techniques, for instance, Bayesian Optimization and Genetic Algorithms. Alongside theoretical explanations, a practical example including Python code as well as the latest research applications of hyperparameter tuning in tourism literature will also be provided.