Schur Nets: effectively exploiting local structure for equivariance in higher order graph neural networks

Neural Information Processing Systems 

GNN operations need to be equivariant to is the starting point for the rest of this paper.