Machine Learning: Finding the Next Superhard Material through Ensemble Learning (Adv. Mater. 5/2021)

In article number 2005112, Jakoah Brgoch and co-workers establish an ensemble machine-learning method to find new superhard materials. The model is trained on the sparse experimental data available in literature predict load-dependent Vickers hardness based only chemical composition. Crystal-structure databases unknown phase diagrams are then screened search of novel materials with outstanding mechanical response.