Warm-starting PDE solvers with any-dimensional machine learning
stat.ML, cs.LG
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/WilsonGregory/ginjax
Terminology
Sources
- A Galois theorem for machine learning: Functions on symmetric matrices and point clouds via lightweight invariant features
- Representation stability in cohomology and asymptotics for families of varieties over finite fields
- Invariant Kernels: Rank Stabilization and Generalization Across Dimensions
- e3nn: Euclidean Neural Networks
- N-body Networks: a Covariant Hierarchical Neural Network Architecture for Learning Atomic Potentials
- Dimension-Free Descriptions of Convex Sets
- Any-Dimensional Polynomial Optimization via de Finetti Theorems
- Any-Dimensional Learning by Sampling
- Fourier Neural Operator for Parametric Partial Differential Equations
- $\mu$pscaling small models: Principled warm starts and hyperparameter transfer
- Towards Coordinate- and Dimension-Agnostic Machine Learning for Partial Differential Equations
- Families of Markov chains with compatible symmetric-group actions
- Some Might Say All You Need Is Sum
- Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds
- Any-Dimensional Invariant Universality
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
- One Permutation Is All You Need: Fast, Deterministic Feature Importance and Model Stress-Testing
- Online Conformal Prediction for Non-Exchangeable Panel Data
- Deep Time-Series Forecasting in 10 Years: A Survey