On the Expressive Power of Sparse Geometric MPNNs
Quick summary
arXiv:2407.02025v5 Announce Type: replace-cross Abstract: Motivated by applications in chemistry and other sciences, we study the expressive power of message-passing neural networks for geometric graphs, whose node features correspond to 3-dimensional positions. Recent work has shown that such models can separate generic pairs of non-isomorphic geometric graphs, though they may fail to separate some rare and complicated instances. However, these results assume a fully connected graph, where each node possesses complete knowledge of all other nodes. In contrast, often, in application, every nod
Key takeaways
- arXiv:2407.02025v5 Announce Type: replace-cross Abstract: Motivated by applications in chemistry and other sciences, we study the expressive power of message-passing neural networks for geometric graphs, whose node features correspond to 3-dimensional positions.
- Recent work has shown that such models can separate generic pairs of non-isomorphic geometric graphs, though they may fail to separate some rare and complicated instances.
- However, these results assume a fully connected graph, where each node possesses complete knowledge of all other nodes.
Why it matters
“On the Expressive Power of Sparse Geometric MPNNs” highlights the need for repeatable measurement rather than a single impressive demonstration. Independent validation across datasets and clearly stated limitations determine whether a result can guide product decisions.
Member comments