Grain boundary slip transfer classification and metric selection with artificial neural networks

Abstract: An artificial neural network is used to evaluate the effectiveness of six metrics and their combinations to assess whether slip transfers across grain boundaries in coarse-grained oligocrystalline Al foils \citep{Bieler_etal2019_2,Alizadeh_etal2020}. This approach extends the one- or two-dimensional projections formerly applied to analyze slip transfer. The accuracy of this binary classification reaches around \pcnt{87} for the best single metric and around \pcnt{90} when considering two or more metrics simultaneously. The results suggest slip transfer mostly depends on the geometric relationship between grains. Training a double-layer network having \num{10} nodes per hidden layer with \num{40} measurements is sufficient to render the maximum accuracy.