Neural ensemble Kalman filter: Data assimilation for compressible flows with shocks
physics.flu-dyn, cs.LG
Submitted: 2026-02-26
Updated: 2026-09-17
Code: https://github.com/kevinwgy/m2c
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Toward a unified data-driven turbulence model through multi-objective learning
- Structurally informed data assimilation in two dimensions
- Sequential estimation of disturbed aerodynamic flows from sparse measurements via a reduced latent space
- Latent Autoencoder Ensemble Kalman Filter for Nonlinear Data assimilation
- BI-EqNO: Generalized Approximate Bayesian Inference with an Equivariant Neural Operator Framework
- Adam: A Method for Stochastic Optimization