A deep BSDE neural network method approximates unnormalized filtering densities for nonlinear Bayesian filtering, trained offline and applied online, with a hybrid a priori-a posteriori error bound proved under the parabolic Hörmander condition.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
math.NA 3representative citing papers
A convergent deep splitting scheme approximates the nonlinear filtering density via Fokker-Planck prediction and exact Bayesian update, with sampling to address high dimensions.
The logarithmic deep backward SDE filter succeeds in a 100-dimensional Lorenz-96 model where particle and ensemble Kalman filters fail, while cutting inference time by two to five orders of magnitude.
citing papers explorer
-
Nonlinear filtering based on density approximation and deep BSDE prediction
A deep BSDE neural network method approximates unnormalized filtering densities for nonlinear Bayesian filtering, trained offline and applied online, with a hybrid a priori-a posteriori error bound proved under the parabolic Hörmander condition.
-
A convergent scheme for the Bayesian filtering problem based on the Fokker--Planck equation and deep splitting
A convergent deep splitting scheme approximates the nonlinear filtering density via Fokker-Planck prediction and exact Bayesian update, with sampling to address high dimensions.
-
High-dimensional Bayesian filtering through deep density approximation
The logarithmic deep backward SDE filter succeeds in a 100-dimensional Lorenz-96 model where particle and ensemble Kalman filters fail, while cutting inference time by two to five orders of magnitude.