Machine Learning Pipelines: Provenance, Reproducibility and FAIR Data Principles

Machine learning (ML) is an increasingly important scientific tool supporting decision making and knowledge generation in numerous fields. With this, it also becomes more that the results of ML experiments are reproducible. Unfortunately, often not case. Rather, ML, similar to many other disciplines, faces a reproducibility crisis. In this paper, we describe our goals initial steps end-to-end pipelines. We investigate which factors beyond availability source code datasets influence experiments. propose ways apply FAIR data practices workflows. present preliminary on role tool, ProvBook, capturing comparing provenance their using Jupyter Notebooks. ReproduceMeGit analyze pipelines described