“…The machine learning approach can overcome the limitation of empirical solutions and preferably describe the nonlinear relationships between high dimensional data. As such, many investigators have taken advantage of advanced machine learning methods to solve complex problems such as the creep rupture life prediction, 23,43 life prediction of components at high temperatures under creep, fatigue, and creep-fatigue conditions, 49 creepfatigue damage assessment, 12 multiaxial fatigue life prediction, 42,45,46,51 FCGR prediction, 24 prediction of mechanical strength of concrete, 28 and so forth. For instance, Zhou et al 51 proposed a machine learning framework that combines the artificial neural network (ANN) and partial least squares (PLS) algorithm to predict fatigue life and identify genetic features of 316LN steel under uniaxial and multiaxial loading conditions.…”