• Corpus ID: 248118511

Ensemble Variational Fokker-Planck Methods for Data Assimilation

@inproceedings{Subrahmanya2021EnsembleVF,
  title={Ensemble Variational Fokker-Planck Methods for Data Assimilation},
  author={Amit N. Subrahmanya and Andrey A. Popov and Adrian Sandu},
  year={2021}
}
. Particle flow filters that smoothly transform particles from being samples of a prior distribution to samples of a posterior, are a major topic of active research. In this work, we introduce a generalized data assimilation framework called the Variational Fokker-Planck method for filtering and smoothing whose specific implementations are the previously known methods such as the mapping particle filter and Langevin-based filters. Using the properties of an optimal Itˆo process that drives the… 

Figures and Tables from this paper