Principled Simplicial Neural Networks for Trajectory Prediction

              T. Mitchell Roddenberry * 1 Nicholas Glaze * 1 Santiago Segarra 1

             Abstract                in their ability to incorporate arbitrary pairwise relational
                                structures in their computations.
   We consider the construction of neural network
   architectures for data on simplicial complexes. In    However, not all data is adequately expressed in terms of
   studying maps on the  ...