StateMixNN learns particle-filter transition and proposal densities as Gaussian mixtures parameterized by neural networks, trained only on the observation likelihood, and reports improved state recovery on Lorenz 96 and Kuramoto systems.
A survey of recent advances in particle filters and remaining challenges for multitarget tracking
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Learning state and proposal dynamics in state-space models using differentiable particle filters and neural networks
StateMixNN learns particle-filter transition and proposal densities as Gaussian mixtures parameterized by neural networks, trained only on the observation likelihood, and reports improved state recovery on Lorenz 96 and Kuramoto systems.