| Abstract: |
| Graph neural networks (GNNs) -- learn graph representations by exploiting graph`s sparsity, connectivity, and symmetries -- have become indispensable for learning geometric data like molecules. However, the most used graphs (e.g., radial cutoff graphs) in molecular modeling lack theoretical guarantees for achieving connectivity and sparsity simultaneously, which are essential for the performance and scalability of GNNs. Furthermore, existing widely used graph construction methods for molecules lack rigidity, limiting GNNs` ability to exploit graph nodes` spatial arrangement. We will present a new hyperparameter-free graph construction of molecules and beyond with sparsity, connectivity, and rigidity guarantees. |
|