Reproducibility in Machine Learning for Health
This work addresses reproducibility issues in machine learning for health, which is crucial for safe and reliable deployment, but it is incremental as it builds on existing concerns and proposes recommendations.
The authors systematically evaluated over 100 recent ML4H papers and found that the field performs poorly in reproducibility, especially regarding data and code accessibility, compared to other machine learning fields.
Machine learning algorithms designed to characterize, monitor, and intervene on human health (ML4H) are expected to perform safely and reliably when operating at scale, potentially outside strict human supervision. This requirement warrants a stricter attention to issues of reproducibility than other fields of machine learning. In this work, we conduct a systematic evaluation of over 100 recently published ML4H research papers along several dimensions related to reproducibility. We find that the field of ML4H compares poorly to more established machine learning fields, particularly concerning data and code accessibility. Finally, drawing from success in other fields of science, we propose recommendations to data providers, academic publishers, and the ML4H research community in order to promote reproducible research moving forward.