The Twisted-Path Particle Filter parameterizes twisting functions via neural networks and optimizes them against a path-measure KL divergence to improve continuous-time particle filtering.
On a strongly convex approximation of a stochastic optimal control problem for importance sampling of metastable diffusions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.CO 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Guidance for twisted particle filter: a continuous-time perspective
The Twisted-Path Particle Filter parameterizes twisting functions via neural networks and optimizes them against a path-measure KL divergence to improve continuous-time particle filtering.