Equivariant Message Passing for the Prediction of Tensorial Properties and
               Molecular Spectra

                Kristof T. Schutt 1 2 Oliver T. Unke 1 2 Michael Gastegger 1 3

              Abstract               ory quickly become prohibitive for large systems and the
                                 prediction of accurate spectra.
    Message passing neural networks have become a
    method of choice for learning on graphs, in par-     Recently, machine learning potentials (Behle ...