Description
Hyperparameters of machine learning algorithms are instrumental for the generation of high-performing models. The tedious task of hyperparamater optimization is nonetheless often reduced to manual optimization or simple methods, a situation often leading to results that are highly sensitive to hyperparameters. In this Talk, we will walk machine learning practitioners through guidelines for efficient hyperparameter optimization based on the framework Oríon. We will present practical approaches for the design of the search space, provide guidelines to select hyperparameter optimization algorithms, and demonstrate how to leverage the pioneering Experiment Version Control provided by Oríon for more efficiency.