Dynamical and Optimization Trade-offs of Levi--Civita Coordinates for Learned Close-Encounter Dynamics
Abhishek Shankar
physics.comp-ph, astro-ph.EP, cs.LG
Submitted: 2026-07-22
Comments: 16 pages, 4 figures
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Hamiltonian Graph Networks with ODE Integrators
- SympNets: Intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems
- Time-adaptive H'enonNets for separable Hamiltonian systems
- Nonseparable Symplectic Neural Networks
- Unifying physical systems' inductive biases in neural ODE using dynamics constraints
- Deep Learning Based Dynamics Identification and Linearization of Orbital Problems using Koopman Theory
- Neural Network Perturbation Theory (NNPT): Learning Residual Corrections from Exact Solutions
- ATLAS-NN: Adaptive Transfer Learnable Symplectic-aware Neural Network for Long-Time Hamiltonian Dynamics
Related papers
- Refining Machine Learning Potentials through Thermodynamic Theory of Phase Transitions
- Polyconvex neural-network surrogates with learned paired directional features for anisotropic hyperelastic homogenization
- Learning Spectrally Optimised Mesh-Free Discretisations
- Feature-preserving Latent-EnKF for Data Assimilation of Flows with Shocks
- chemtrain-deploy: A parallel and scalable framework for machine learning potentials in million-atom MD simulations
- Omnisolver: An extensible interface to Ising spin-glass and QUBO solvers: adding a distributed GPU brute-force plugin