Matches in SemOpenAlex for { <https://semopenalex.org/work/W2263340393> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W2263340393 abstract "Applying standard Markov chain Monte Carlo (MCMC) algorithms to large data sets is computationally infeasible. The recently proposed stochastic gradient Langevin dynamics (SGLD) method circumvents this problem in three ways: it generates proposed moves using only a subset of the data, it skips the Metropolis-Hastings accept-reject step, and it uses sequences of decreasing step sizes. In cite{TehThierryVollmerSGLD2014}, we provided the mathematical foundations for the decreasing step size SGLD, including consistency and a central limit theorem. However, in practice the SGLD is run for a relatively small number of iterations, and its step size is not decreased to zero. The present article investigates the behaviour of the SGLD with fixed step size. In particular we characterise the asymptotic bias explicitly, along with its dependence on the step size and the variance of the stochastic gradient. On that basis a modified SGLD which removes the asymptotic bias due to the variance of the stochastic gradients up to first order in the step size is derived. Moreover, we are able to obtain bounds on the finite-time bias, variance and mean squared error (MSE). The theory is illustrated with a Gaussian toy model for which the bias and the MSE for the estimation of moments can be obtained explicitly. For this toy model we study the gain of the SGLD over the standard Euler method in the limit of large data sets." @default.
- W2263340393 created "2016-06-24" @default.
- W2263340393 creator A5003185028 @default.
- W2263340393 creator A5024728732 @default.
- W2263340393 creator A5068361059 @default.
- W2263340393 date "2015-01-02" @default.
- W2263340393 modified "2023-09-27" @default.
- W2263340393 title "(Non-) asymptotic properties of Stochastic Gradient Langevin Dynamics" @default.
- W2263340393 cites W2951105989 @default.
- W2263340393 hasPublicationYear "2015" @default.
- W2263340393 type Work @default.
- W2263340393 sameAs 2263340393 @default.
- W2263340393 citedByCount "24" @default.
- W2263340393 countsByYear W22633403932015 @default.
- W2263340393 countsByYear W22633403932016 @default.
- W2263340393 countsByYear W22633403932017 @default.
- W2263340393 countsByYear W22633403932018 @default.
- W2263340393 countsByYear W22633403932019 @default.
- W2263340393 countsByYear W22633403932020 @default.
- W2263340393 crossrefType "posted-content" @default.
- W2263340393 hasAuthorship W2263340393A5003185028 @default.
- W2263340393 hasAuthorship W2263340393A5024728732 @default.
- W2263340393 hasAuthorship W2263340393A5068361059 @default.
- W2263340393 hasConcept C105795698 @default.
- W2263340393 hasConcept C111350023 @default.
- W2263340393 hasConcept C118615104 @default.
- W2263340393 hasConcept C121332964 @default.
- W2263340393 hasConcept C121864883 @default.
- W2263340393 hasConcept C134306372 @default.
- W2263340393 hasConcept C151201525 @default.
- W2263340393 hasConcept C185429906 @default.
- W2263340393 hasConcept C19499675 @default.
- W2263340393 hasConcept C2776436953 @default.
- W2263340393 hasConcept C2780004032 @default.
- W2263340393 hasConcept C28826006 @default.
- W2263340393 hasConcept C33923547 @default.
- W2263340393 hasConcept C971699 @default.
- W2263340393 hasConceptScore W2263340393C105795698 @default.
- W2263340393 hasConceptScore W2263340393C111350023 @default.
- W2263340393 hasConceptScore W2263340393C118615104 @default.
- W2263340393 hasConceptScore W2263340393C121332964 @default.
- W2263340393 hasConceptScore W2263340393C121864883 @default.
- W2263340393 hasConceptScore W2263340393C134306372 @default.
- W2263340393 hasConceptScore W2263340393C151201525 @default.
- W2263340393 hasConceptScore W2263340393C185429906 @default.
- W2263340393 hasConceptScore W2263340393C19499675 @default.
- W2263340393 hasConceptScore W2263340393C2776436953 @default.
- W2263340393 hasConceptScore W2263340393C2780004032 @default.
- W2263340393 hasConceptScore W2263340393C28826006 @default.
- W2263340393 hasConceptScore W2263340393C33923547 @default.
- W2263340393 hasConceptScore W2263340393C971699 @default.
- W2263340393 hasLocation W22633403931 @default.
- W2263340393 hasOpenAccess W2263340393 @default.
- W2263340393 hasPrimaryLocation W22633403931 @default.
- W2263340393 hasRelatedWork W123701566 @default.
- W2263340393 hasRelatedWork W1545319692 @default.
- W2263340393 hasRelatedWork W1624701674 @default.
- W2263340393 hasRelatedWork W1880262756 @default.
- W2263340393 hasRelatedWork W1973794074 @default.
- W2263340393 hasRelatedWork W1983452151 @default.
- W2263340393 hasRelatedWork W1994616650 @default.
- W2263340393 hasRelatedWork W2059448777 @default.
- W2263340393 hasRelatedWork W2110529144 @default.
- W2263340393 hasRelatedWork W2144193737 @default.
- W2263340393 hasRelatedWork W2165652770 @default.
- W2263340393 hasRelatedWork W2167433878 @default.
- W2263340393 hasRelatedWork W2186210550 @default.
- W2263340393 hasRelatedWork W2238987678 @default.
- W2263340393 hasRelatedWork W2478027467 @default.
- W2263340393 hasRelatedWork W2590516361 @default.
- W2263340393 hasRelatedWork W2947548899 @default.
- W2263340393 hasRelatedWork W2951266961 @default.
- W2263340393 hasRelatedWork W3101060993 @default.
- W2263340393 hasRelatedWork W3184468133 @default.
- W2263340393 isParatext "false" @default.
- W2263340393 isRetracted "false" @default.
- W2263340393 magId "2263340393" @default.
- W2263340393 workType "article" @default.