Symplectic Normal Coordinates Flow for Generative Modeling of Hamiltonian Systems
physics.acc-ph, cs.LG
Submitted: 2026-11-08
Updated: 2026-11-08
Code: https://github.com/electronsandstuff/Normal-Coordinates-Flow
Terminology
Sources
- Symplectic Generative Networks (SGNs): A Hamiltonian Framework for Invertible Deep Generative Modeling
- Learning Symmetries of Classical Integrable Systems
- Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
- Fast neural Poincar'e maps for toroidal magnetic fields
- Gradient Networks
- Learning Integrable Dynamics with Action-Angle Networks
- Density estimation using Real NVP
- Optimal unit triangular factorization of symplectic matrices
- SympNets: Intrinsic structure-preserving symplectic networks for identifying Hamiltonian systems
- Adam: A Method for Stochastic Optimization
- SGDR: Stochastic Gradient Descent with Warm Restarts
- Equivariant Hamiltonian Flows
- Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
- Nonseparable Symplectic Neural Networks
- Normalizing Flows are Capable Generative Models
Related papers
- Design and Operation of a 3-Crystal Multi-Volume Reflection Array for Loss Reduction in CERN SPS Slow Extraction
- Booster-based beam recycling for swap-out injection at the High Energy Photon Source
- Design and study of the electron beam-line at the European XFEL (ELBEX) for strong field quantum electro-dynamics
- Comments on the paper "Eliminating beam-induced depolarizing effects in the hydrogen jet target for high-precision proton beam polarimetry at the Electron-Ion Collider"
- Impedance of multilayer cylindrical structures with a material-filled beam region
- Polarization-dependent exchange effects in intrabeam scattering