architecture · first seen 16 Oct 2026
SO(3)-equivariant neural network
Also: spherical CNN, SO(2)-equivariant layers
Neural networks that are equivariant to 3D rotations, applied to data defined on the sphere.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the spherePapers · 16 Oct 2026 · Joshua Williamson, Niall Jeffrey, Lorne Whiteway
- Accelerating SO(2)-Equivariant Materials Models through Low-Precision EmulationPapers · 26 Oct 2026 · Denghui Lu, Alexander Maeder, Alexandros Nikolaos Ziogas, Mathieu Luisier, Manas…
- SO(3)-Equivariant Learning on CAD Boundary RepresentationsPapers · 3 Nov 2026 · Matteo Ballegeer, Dries F. Benoit
Latest sources
- SO(3)-Equivariant Learning on CAD Boundary RepresentationsPapers · 3 Nov 2026 · Matteo Ballegeer, Dries F. Benoit
- Accelerating SO(2)-Equivariant Materials Models through Low-Precision EmulationPapers · 26 Oct 2026 · Denghui Lu, Alexander Maeder, Alexandros Nikolaos Ziogas, Mathieu Luisier, Manas…
- Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the spherePapers · 16 Oct 2026 · Joshua Williamson, Niall Jeffrey, Lorne Whiteway