Abstract: The performance of machine learning models in real-world applications is often challenged by data drift, in which the statistical properties of the data evolve over time. This phenomenon is ...