E (3)-Equivariant Mesh Neural Networks

TA Trang, NK Ngo, DT Levy, TN Vo… - International …, 2024 - proceedings.mlr.press
Triangular meshes are widely used to represent three-dimensional objects. As a result,
many recent works have addressed the need for geometric deep learning on 3D meshes …

Weisfeiler Leman for Euclidean Equivariant Machine Learning

S Hordan, T Amir, N Dym - arxiv preprint arxiv:2402.02484, 2024 - arxiv.org
The $ k $-Weifeiler-Leman ($ k $-WL) graph isomorphism test hierarchy is a common
method for assessing the expressive power of graph neural networks (GNNs). Recently, the …

On the Expressive Power of Sparse Geometric MPNNs

Y Sverdlov, N Dym - arxiv preprint arxiv:2407.02025, 2024 - arxiv.org
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 …