Matches in SemOpenAlex for { <https://semopenalex.org/work/W2953377880> ?p ?o ?g. }
- W2953377880 abstract "Because of their multimodality, mixture posterior distributions are difficult to sample with standard Markov chain Monte Carlo (MCMC) methods. We propose a strategy to enhance the sampling of MCMC in this context, using a biasing procedure which originates from computational Statistical Physics. The principle is first to choose a coordinate, that is, a direction in which the target distribution is multimodal. In a second step, the marginal log-density of the reaction coordinate with respect to the posterior distribution is estimated; minus this quantity is called free energy in the computational Statistical Physics literature. To this end, we use adaptive biasing Markov chain algorithms which adapt their targeted invariant distribution on the fly, in order to overcome sampling barriers along the chosen reaction coordinate. Finally, we perform an importance sampling step in order to remove the bias and recover the true posterior. The efficiency factor of the importance sampling step can easily be estimated emph{a priori} once the bias is known, and appears to be rather large for the test cases we considered. A crucial point is the choice of the reaction coordinate. One standard choice (used for example in the classical Wang-Landau algorithm) is minus the log-posterior density. We discuss other choices. We show in particular that the hyper-parameter that determines the order of magnitude of the variance of each component is both a convenient and an efficient reaction coordinate. We also show how to adapt the method to compute the evidence (marginal likelihood) of a mixture model. We illustrate our approach by analyzing two real data sets." @default.
- W2953377880 created "2019-06-27" @default.
- W2953377880 creator A5015347592 @default.
- W2953377880 creator A5041602248 @default.
- W2953377880 creator A5049392160 @default.
- W2953377880 date "2010-03-01" @default.
- W2953377880 modified "2023-10-10" @default.
- W2953377880 title "Free Energy Methods for Bayesian Inference: Efficient Exploration of Univariate Gaussian Mixture Posteriors" @default.
- W2953377880 cites W1480714962 @default.
- W2953377880 cites W1513873506 @default.
- W2953377880 cites W1579271636 @default.
- W2953377880 cites W1593038947 @default.
- W2953377880 cites W1602551265 @default.
- W2953377880 cites W1616675165 @default.
- W2953377880 cites W1970897635 @default.
- W2953377880 cites W1995907662 @default.
- W2953377880 cites W2003582562 @default.
- W2953377880 cites W2013164703 @default.
- W2953377880 cites W2014574246 @default.
- W2953377880 cites W2019486793 @default.
- W2953377880 cites W2027445478 @default.
- W2953377880 cites W2029186981 @default.
- W2953377880 cites W2035756456 @default.
- W2953377880 cites W2041694664 @default.
- W2953377880 cites W2042515509 @default.
- W2953377880 cites W2056760934 @default.
- W2953377880 cites W2057487175 @default.
- W2953377880 cites W2078629389 @default.
- W2953377880 cites W2078644085 @default.
- W2953377880 cites W2096878708 @default.
- W2953377880 cites W2135973421 @default.
- W2953377880 cites W2148444003 @default.
- W2953377880 cites W2166698530 @default.
- W2953377880 cites W2963276153 @default.
- W2953377880 cites W3101769934 @default.
- W2953377880 cites W605692607 @default.
- W2953377880 cites W610242975 @default.
- W2953377880 hasPublicationYear "2010" @default.
- W2953377880 type Work @default.
- W2953377880 sameAs 2953377880 @default.
- W2953377880 citedByCount "2" @default.
- W2953377880 crossrefType "posted-content" @default.
- W2953377880 hasAuthorship W2953377880A5015347592 @default.
- W2953377880 hasAuthorship W2953377880A5041602248 @default.
- W2953377880 hasAuthorship W2953377880A5049392160 @default.
- W2953377880 hasConcept C105795698 @default.
- W2953377880 hasConcept C106131492 @default.
- W2953377880 hasConcept C107673813 @default.
- W2953377880 hasConcept C111350023 @default.
- W2953377880 hasConcept C11413529 @default.
- W2953377880 hasConcept C121332964 @default.
- W2953377880 hasConcept C122123141 @default.
- W2953377880 hasConcept C140779682 @default.
- W2953377880 hasConcept C147597530 @default.
- W2953377880 hasConcept C160234255 @default.
- W2953377880 hasConcept C161584116 @default.
- W2953377880 hasConcept C163716315 @default.
- W2953377880 hasConcept C165216359 @default.
- W2953377880 hasConcept C170593435 @default.
- W2953377880 hasConcept C18553476 @default.
- W2953377880 hasConcept C185592680 @default.
- W2953377880 hasConcept C199163554 @default.
- W2953377880 hasConcept C31972630 @default.
- W2953377880 hasConcept C33923547 @default.
- W2953377880 hasConcept C41008148 @default.
- W2953377880 hasConcept C57830394 @default.
- W2953377880 hasConcept C62520636 @default.
- W2953377880 hasConcept C95923904 @default.
- W2953377880 hasConcept C98763669 @default.
- W2953377880 hasConceptScore W2953377880C105795698 @default.
- W2953377880 hasConceptScore W2953377880C106131492 @default.
- W2953377880 hasConceptScore W2953377880C107673813 @default.
- W2953377880 hasConceptScore W2953377880C111350023 @default.
- W2953377880 hasConceptScore W2953377880C11413529 @default.
- W2953377880 hasConceptScore W2953377880C121332964 @default.
- W2953377880 hasConceptScore W2953377880C122123141 @default.
- W2953377880 hasConceptScore W2953377880C140779682 @default.
- W2953377880 hasConceptScore W2953377880C147597530 @default.
- W2953377880 hasConceptScore W2953377880C160234255 @default.
- W2953377880 hasConceptScore W2953377880C161584116 @default.
- W2953377880 hasConceptScore W2953377880C163716315 @default.
- W2953377880 hasConceptScore W2953377880C165216359 @default.
- W2953377880 hasConceptScore W2953377880C170593435 @default.
- W2953377880 hasConceptScore W2953377880C18553476 @default.
- W2953377880 hasConceptScore W2953377880C185592680 @default.
- W2953377880 hasConceptScore W2953377880C199163554 @default.
- W2953377880 hasConceptScore W2953377880C31972630 @default.
- W2953377880 hasConceptScore W2953377880C33923547 @default.
- W2953377880 hasConceptScore W2953377880C41008148 @default.
- W2953377880 hasConceptScore W2953377880C57830394 @default.
- W2953377880 hasConceptScore W2953377880C62520636 @default.
- W2953377880 hasConceptScore W2953377880C95923904 @default.
- W2953377880 hasConceptScore W2953377880C98763669 @default.
- W2953377880 hasLocation W29533778801 @default.
- W2953377880 hasOpenAccess W2953377880 @default.
- W2953377880 hasPrimaryLocation W29533778801 @default.
- W2953377880 hasRelatedWork W1536565998 @default.
- W2953377880 hasRelatedWork W1788616518 @default.
- W2953377880 hasRelatedWork W2011190924 @default.
- W2953377880 hasRelatedWork W2012656342 @default.