2020
DOI: 10.48550/arxiv.2009.03699
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Accelerating sequential Monte Carlo with surrogate likelihoods

Abstract: Delayed-acceptance is a technique for reducing computational effort for Bayesian models with expensive likelihoods. Using a delayed-acceptance kernel for Markov chain Monte Carlo can reduce the number of expensive likelihoods evaluations required to approximate a posterior expectation. Delayed-acceptance uses a surrogate, or approximate, likelihood to avoid evaluation of the expensive likelihood when possible. Within the sequential Monte Carlo framework, we utilise the history of the sampler to adaptively tune… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
2
0

Year Published

2021
2021
2021
2021

Publication Types

Select...
2

Relationship

0
2

Authors

Journals

citations
Cited by 2 publications
(2 citation statements)
references
References 35 publications
0
2
0
Order By: Relevance
“…Recently, there has been substantial research activity in the application of approximate model simulations and posterior samplers in combination with bias correction adjustments to accelerate likelihood-free applications that would be impractical otherwise. For example, delayed-acceptance [35,36,37,38] or early-rejection approaches [39]. Techniques such as transport maps [40] and moment-matching transforms [41] aim to transform a set of approximate posterior samples, using a surrogate or reduced model, into posterior samples under an expensive exact model.…”
Section: Introductionmentioning
confidence: 99%
“…Recently, there has been substantial research activity in the application of approximate model simulations and posterior samplers in combination with bias correction adjustments to accelerate likelihood-free applications that would be impractical otherwise. For example, delayed-acceptance [35,36,37,38] or early-rejection approaches [39]. Techniques such as transport maps [40] and moment-matching transforms [41] aim to transform a set of approximate posterior samples, using a surrogate or reduced model, into posterior samples under an expensive exact model.…”
Section: Introductionmentioning
confidence: 99%
“…We focus mainly on the static batch scenario for MCMC and IS algorithms. However, most of the results presented in this work can be extended to the sequential framework (consider, e.g., the recent work of [13]). We classify the studied techniques in different families, and provide several explanatory tables and figures.…”
Section: Introductionmentioning
confidence: 99%