Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100739774> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W3100739774 endingPage "9097" @default.
- W3100739774 startingPage "9086" @default.
- W3100739774 abstract "Gaussian latent variable models are a key class of Bayesian hierarchical models with applications in many fields. Performing Bayesian inference on such models can be challenging as Markov chain Monte Carlo algorithms struggle with the geometry of the resulting posterior distribution and can be prohibitively slow. An alternative is to use a Laplace approximation to marginalize out the latent Gaussian variables and then integrate out the remaining hyperparameters using dynamic Hamiltonian Monte Carlo, a gradient-based Markov chain Monte Carlo sampler. To implement this scheme efficiently, we derive a novel adjoint method that propagates the minimal information needed to construct the gradient of the approximate marginal likelihood. This strategy yields a scalable differentiation method that is orders of magnitude faster than state of the art differentiation techniques when the hyperparameters are high dimensional. We prototype the method in the probabilistic programming framework Stan and test the utility of the embedded Laplace approximation on several models, including one where the dimension of the hyperparameter is $sim$6,000. Depending on the cases, the benefits can include an alleviation of the geometric pathologies that frustrate Hamiltonian Monte Carlo and a dramatic speed-up." @default.
- W3100739774 created "2020-11-23" @default.
- W3100739774 creator A5030390434 @default.
- W3100739774 creator A5041115861 @default.
- W3100739774 creator A5047269552 @default.
- W3100739774 creator A5063283958 @default.
- W3100739774 date "2020-01-01" @default.
- W3100739774 modified "2023-09-24" @default.
- W3100739774 title "Hamiltonian Monte Carlo using an adjoint-differentiated Laplace approximation: Bayesian inference for latent Gaussian models and beyond" @default.
- W3100739774 hasPublicationYear "2020" @default.
- W3100739774 type Work @default.
- W3100739774 sameAs 3100739774 @default.
- W3100739774 citedByCount "6" @default.
- W3100739774 countsByYear W31007397742020 @default.
- W3100739774 countsByYear W31007397742021 @default.
- W3100739774 crossrefType "proceedings-article" @default.
- W3100739774 hasAuthorship W3100739774A5030390434 @default.
- W3100739774 hasAuthorship W3100739774A5041115861 @default.
- W3100739774 hasAuthorship W3100739774A5047269552 @default.
- W3100739774 hasAuthorship W3100739774A5063283958 @default.
- W3100739774 hasConcept C105795698 @default.
- W3100739774 hasConcept C107673813 @default.
- W3100739774 hasConcept C111350023 @default.
- W3100739774 hasConcept C11413529 @default.
- W3100739774 hasConcept C121332964 @default.
- W3100739774 hasConcept C121864883 @default.
- W3100739774 hasConcept C126255220 @default.
- W3100739774 hasConcept C13153151 @default.
- W3100739774 hasConcept C154945302 @default.
- W3100739774 hasConcept C160234255 @default.
- W3100739774 hasConcept C169272836 @default.
- W3100739774 hasConcept C19499675 @default.
- W3100739774 hasConcept C22243797 @default.
- W3100739774 hasConcept C2776214188 @default.
- W3100739774 hasConcept C28826006 @default.
- W3100739774 hasConcept C33923547 @default.
- W3100739774 hasConcept C41008148 @default.
- W3100739774 hasConcept C51167844 @default.
- W3100739774 hasConcept C8642999 @default.
- W3100739774 hasConceptScore W3100739774C105795698 @default.
- W3100739774 hasConceptScore W3100739774C107673813 @default.
- W3100739774 hasConceptScore W3100739774C111350023 @default.
- W3100739774 hasConceptScore W3100739774C11413529 @default.
- W3100739774 hasConceptScore W3100739774C121332964 @default.
- W3100739774 hasConceptScore W3100739774C121864883 @default.
- W3100739774 hasConceptScore W3100739774C126255220 @default.
- W3100739774 hasConceptScore W3100739774C13153151 @default.
- W3100739774 hasConceptScore W3100739774C154945302 @default.
- W3100739774 hasConceptScore W3100739774C160234255 @default.
- W3100739774 hasConceptScore W3100739774C169272836 @default.
- W3100739774 hasConceptScore W3100739774C19499675 @default.
- W3100739774 hasConceptScore W3100739774C22243797 @default.
- W3100739774 hasConceptScore W3100739774C2776214188 @default.
- W3100739774 hasConceptScore W3100739774C28826006 @default.
- W3100739774 hasConceptScore W3100739774C33923547 @default.
- W3100739774 hasConceptScore W3100739774C41008148 @default.
- W3100739774 hasConceptScore W3100739774C51167844 @default.
- W3100739774 hasConceptScore W3100739774C8642999 @default.
- W3100739774 hasLocation W31007397741 @default.
- W3100739774 hasOpenAccess W3100739774 @default.
- W3100739774 hasPrimaryLocation W31007397741 @default.
- W3100739774 hasRelatedWork W1655728402 @default.
- W3100739774 hasRelatedWork W2034338630 @default.
- W3100739774 hasRelatedWork W2085893390 @default.
- W3100739774 hasRelatedWork W2156520836 @default.
- W3100739774 hasRelatedWork W2219657383 @default.
- W3100739774 hasRelatedWork W2311351094 @default.
- W3100739774 hasRelatedWork W2346910172 @default.
- W3100739774 hasRelatedWork W2522620351 @default.
- W3100739774 hasRelatedWork W2572702041 @default.
- W3100739774 hasRelatedWork W2577537660 @default.
- W3100739774 hasRelatedWork W2903618527 @default.
- W3100739774 hasRelatedWork W2953167674 @default.
- W3100739774 hasRelatedWork W2962801292 @default.
- W3100739774 hasRelatedWork W2963218199 @default.
- W3100739774 hasRelatedWork W2963348289 @default.
- W3100739774 hasRelatedWork W2977844912 @default.
- W3100739774 hasRelatedWork W3019228676 @default.
- W3100739774 hasRelatedWork W3033831365 @default.
- W3100739774 hasRelatedWork W3093702125 @default.
- W3100739774 hasRelatedWork W3101380508 @default.
- W3100739774 hasVolume "33" @default.
- W3100739774 isParatext "false" @default.
- W3100739774 isRetracted "false" @default.
- W3100739774 magId "3100739774" @default.
- W3100739774 workType "article" @default.