How to set up your first machine learning project in astronomy

TL;DR

This paper offers practical guidelines for setting up machine learning projects in astronomy to enhance scientific robustness, efficiency, and insight quality, with potential applicability to other scientific fields.

Contribution

It provides a structured set of guidelines for designing and executing machine learning projects in astronomy, emphasizing clarity, workflow, and validation to improve scientific outcomes.

Findings

Guidelines improve project robustness and efficiency

Enhanced clarity leads to better scientific insights

Workflow recommendations reduce frustration and time investment

Abstract

Large, freely available, well-maintained data sets have made astronomy a popular playground for machine learning projects. Nevertheless, robust insights gained to both machine learning and physics could be improved by clarity in problem definition and establishing workflows that critically verify, characterize and calibrate machine learning models. We provide a collection of guidelines to setting up machine learning projects to make them likely useful for science, less frustrating and time-intensive for the scientist and their computers, and more likely to lead to robust insights. We draw examples and experience from astronomy, but the advice is potentially applicable to other areas in science.