Matches in SemOpenAlex for { <https://semopenalex.org/work/W2893924688> ?p ?o ?g. }
- W2893924688 abstract "Bayesian inference typically requires the computation of an approximation to the posterior distribution. An important requirement for an approximate Bayesian inference algorithm is to output high-accuracy posterior mean and uncertainty estimates. Classical Monte Carlo methods, particularly Markov Chain Monte Carlo, remain the gold standard for approximate Bayesian inference because they have a robust finite-sample theory and reliable convergence diagnostics. However, alternative methods, which are more scalable or apply to problems where Markov Chain Monte Carlo cannot be used, lack the same finite-data approximation theory and tools for evaluating their accuracy. In this work, we develop a flexible new approach to bounding the error of mean and uncertainty estimates of scalable inference algorithms. Our strategy is to control the estimation errors in terms of Wasserstein distance, then bound the Wasserstein distance via a generalized notion of Fisher distance. Unlike computing the Wasserstein distance, which requires access to the normalized posterior distribution, the Fisher distance is tractable to compute because it requires access only to the gradient of the log posterior density. We demonstrate the usefulness of our Fisher distance approach by deriving bounds on the Wasserstein error of the Laplace approximation and Hilbert coresets. We anticipate that our approach will be applicable to many other approximate inference methods such as the integrated Laplace approximation, variational inference, and approximate Bayesian computation" @default.
- W2893924688 created "2018-10-05" @default.
- W2893924688 creator A5004323430 @default.
- W2893924688 creator A5025539763 @default.
- W2893924688 creator A5051096462 @default.
- W2893924688 creator A5065653386 @default.
- W2893924688 date "2018-09-25" @default.
- W2893924688 modified "2023-09-27" @default.
- W2893924688 title "Practical bounds on the error of Bayesian posterior approximations: A nonasymptotic approach." @default.
- W2893924688 cites W1480392687 @default.
- W2893924688 cites W1505878979 @default.
- W2893924688 cites W1551558821 @default.
- W2893924688 cites W1624701674 @default.
- W2893924688 cites W1964607942 @default.
- W2893924688 cites W2006593990 @default.
- W2893924688 cites W2018304737 @default.
- W2893924688 cites W2019420194 @default.
- W2893924688 cites W2045656233 @default.
- W2893924688 cites W2045973738 @default.
- W2893924688 cites W2111698756 @default.
- W2893924688 cites W2139812092 @default.
- W2893924688 cites W2144898279 @default.
- W2893924688 cites W2146200992 @default.
- W2893924688 cites W2149498546 @default.
- W2893924688 cites W2167433878 @default.
- W2893924688 cites W2170316986 @default.
- W2893924688 cites W2263340393 @default.
- W2893924688 cites W2277729655 @default.
- W2893924688 cites W2305001871 @default.
- W2893924688 cites W2340515955 @default.
- W2893924688 cites W2404724429 @default.
- W2893924688 cites W2592105405 @default.
- W2893924688 cites W2787219950 @default.
- W2893924688 cites W2887005408 @default.
- W2893924688 cites W2949933384 @default.
- W2893924688 cites W2950204717 @default.
- W2893924688 cites W2951033158 @default.
- W2893924688 cites W2962771927 @default.
- W2893924688 cites W2963173382 @default.
- W2893924688 cites W2963505885 @default.
- W2893924688 cites W2963649795 @default.
- W2893924688 cites W2963888630 @default.
- W2893924688 cites W2964004456 @default.
- W2893924688 cites W2964015990 @default.
- W2893924688 cites W2964282829 @default.
- W2893924688 cites W2964293062 @default.
- W2893924688 cites W2964332587 @default.
- W2893924688 cites W3098512007 @default.
- W2893924688 cites W3106022061 @default.
- W2893924688 cites W3127518054 @default.
- W2893924688 cites W48692188 @default.
- W2893924688 cites W2606251623 @default.
- W2893924688 hasPublicationYear "2018" @default.
- W2893924688 type Work @default.
- W2893924688 sameAs 2893924688 @default.
- W2893924688 citedByCount "10" @default.
- W2893924688 countsByYear W28939246882019 @default.
- W2893924688 countsByYear W28939246882020 @default.
- W2893924688 countsByYear W28939246882021 @default.
- W2893924688 crossrefType "posted-content" @default.
- W2893924688 hasAuthorship W2893924688A5004323430 @default.
- W2893924688 hasAuthorship W2893924688A5025539763 @default.
- W2893924688 hasAuthorship W2893924688A5051096462 @default.
- W2893924688 hasAuthorship W2893924688A5065653386 @default.
- W2893924688 hasConcept C105795698 @default.
- W2893924688 hasConcept C107673813 @default.
- W2893924688 hasConcept C111350023 @default.
- W2893924688 hasConcept C11413529 @default.
- W2893924688 hasConcept C126255220 @default.
- W2893924688 hasConcept C134261354 @default.
- W2893924688 hasConcept C154945302 @default.
- W2893924688 hasConcept C160234255 @default.
- W2893924688 hasConcept C19499675 @default.
- W2893924688 hasConcept C22243797 @default.
- W2893924688 hasConcept C2776214188 @default.
- W2893924688 hasConcept C2777472644 @default.
- W2893924688 hasConcept C2779377595 @default.
- W2893924688 hasConcept C28826006 @default.
- W2893924688 hasConcept C33923547 @default.
- W2893924688 hasConcept C41008148 @default.
- W2893924688 hasConcept C57830394 @default.
- W2893924688 hasConceptScore W2893924688C105795698 @default.
- W2893924688 hasConceptScore W2893924688C107673813 @default.
- W2893924688 hasConceptScore W2893924688C111350023 @default.
- W2893924688 hasConceptScore W2893924688C11413529 @default.
- W2893924688 hasConceptScore W2893924688C126255220 @default.
- W2893924688 hasConceptScore W2893924688C134261354 @default.
- W2893924688 hasConceptScore W2893924688C154945302 @default.
- W2893924688 hasConceptScore W2893924688C160234255 @default.
- W2893924688 hasConceptScore W2893924688C19499675 @default.
- W2893924688 hasConceptScore W2893924688C22243797 @default.
- W2893924688 hasConceptScore W2893924688C2776214188 @default.
- W2893924688 hasConceptScore W2893924688C2777472644 @default.
- W2893924688 hasConceptScore W2893924688C2779377595 @default.
- W2893924688 hasConceptScore W2893924688C28826006 @default.
- W2893924688 hasConceptScore W2893924688C33923547 @default.
- W2893924688 hasConceptScore W2893924688C41008148 @default.
- W2893924688 hasConceptScore W2893924688C57830394 @default.
- W2893924688 hasLocation W28939246881 @default.
- W2893924688 hasOpenAccess W2893924688 @default.