Abstract: We employ physics-informed neural networks (PINNs) to quantify the microstructure of polycrystalline nickel by computing the spatial variation of compliance coefficients (compressibility, ...