This paper demonstrates the use of machine learning to extract material parameters from a single magnetic domain image. Specifically, it focuses on estimating the Dzyaloshinskii-Moriya (DM) interaction and magnetic anisotropy distribution in thin-film heterostructures, crucial for next-generation magnetic memory technologies. A convolutional neural network trained on micromagnetic simulations accurately estimates the DM exchange constant and anisotropy distribution, showing good agreement with experimental results. This approach simplifies experimental processes and broadens the scope of materials research.
