Key structures to metallic glass stability revealed with machine learning

Using the second-nearest neighboring atoms to predict metallic glass stability can help researchers more accurately model the disordered solid with strong, elastic properties.