Setting standards for data driven materials science

This article addresses the growing use of machine learning (ML) in materials science and the need for rigorous standards to ensure high-quality research. It highlights the importance of openness, reproducibility, and proper validation to avoid inflated expectations and unreliable results. The authors introduce a checklist designed to guide the assessment of ML-based studies, emphasizing transparency and data availability. This initiative aims to maintain scientific rigor and adapt to rapid advancements, ensuring the responsible integration of ML in the materials science community.

Please find the link to the complete article on this topic below:

https://www.nature.com/articles/s41524-024-01411-6