Nature

Uncertainty quantification for Bayesian active learning in rupture life prediction of ferritic steels

Advanced ultra-supercritical power plants require increased steam temperature and pressure, for higher efficiency and lower carbon emissions, to comply with environmental regulations. Without building ...
Nature

Best practices for multi-fidelity Bayesian optimization in materials and molecular research

Among machine learning techniques, Bayesian optimization (BO) has emerged as the go-to choice for optimizing the design of experiments in the chemical domain 1,2,3,4. BO, grounded in a probabilistic ...