The occlusion process is a parallel, variational-assisted wrapper that decorrelates MCMC samples and provably inherits LLN, normed convergence, and geometric ergodicity, with empirical variance reduction when the variational proposal is good.
On the notion of recurrence in discrete stochastic processes
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.CO 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The occlusion process: improving sampler performance with parallel computation and variational approximation
The occlusion process is a parallel, variational-assisted wrapper that decorrelates MCMC samples and provably inherits LLN, normed convergence, and geometric ergodicity, with empirical variance reduction when the variational proposal is good.