ML4DM '23: The Third Workshop on the Emerging Applications of Machine Learning in Modern Data Management.

Machine Learning (ML) has gained prominence across various fields, including data management. Rule-based components are substituted by ML-driven counterparts that extract rules from experience. The prevalence of statistical methods is waning as approaches that learn functional dependencies, correlations, and data skewness emerge. Learning-based techniques hold advantages, reducing the cost of developing and maintaining highly complex classical modules while tailoring behavior to individual system needs. This workshop convened leaders of research projects and audiences from academia and industry. The goal was to explore examples of utilizing ML to modernize data management. The discussed instances spanned four categories: Cardinality Estimation, Database Knobs Tuning, Data Partitioning, and Query Plan Evaluation.