An optimal transport reformulation of Bayes' rule lets neural networks approximate posterior samples in nonlinear filtering, with error controlled by the optimization gap.
Freedman , volume 2 , publisher Institute of Mathematical Sciences , pages 316--334
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SY 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
How to implement the Bayes' formula in the age of ML?
An optimal transport reformulation of Bayes' rule lets neural networks approximate posterior samples in nonlinear filtering, with error controlled by the optimization gap.