Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the sphere
stat.ML, astro-ph.CO, cs.LG
Submitted: 2026-10-16
Updated: 2026-10-16
Related papers
- Behavior of prediction performance metrics with rare events
- Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
- A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
- Efficient quadratic entropy with distance sketches
- Embedding-Bias in Conditional Independence Testing
- From Log-Odds to Shapley Values: An Explanatory Geometry for the Weighted Naive Bayes Classifier