Machine Learning Accelerated, High Throughput, Multi‐Objective Optimization of Multiprincipal Element Alloys (Small 42/2021)

Multiprincipal Element Alloys In article number 2102972, Teng Li and co-workers present a highly efficient design strategy of multiprincipal element alloys (MPEAs) through a coherent integration of molecular dynamic simulation, machine learning algorithms and genetic algorithm. Such a design strategy not only yields remarkable precision of prediction of the critical resolved shear stress and Young's modulus of CoNiCrFeMn MPEAs with an impressively low error, but also enables a drastic 12 600-fold reduction of prediction time in comparison with pure atomic simulations. The multi-objective genetic algorithm further helps identify 100 optimal MPEA compositions with both high critical resolved shear stress and high stiffness.