FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963964397> ?p ?o ?g. }

• W2963964397 endingPage "A2487" @default.
• W2963964397 startingPage "A2451" @default.
• W2963964397 abstract "In the Bayesian approach to inverse problems, data are often informative, relative to the prior, only on a low-dimensional subspace of the parameter space. Significant computational savings can be achieved by using this subspace to characterize and approximate the posterior distribution of the parameters. We first investigate approximation of the posterior covariance matrix as a low-rank update of the prior covariance matrix. We prove optimality of a particular update, based on the leading eigendirections of the matrix pencil defined by the Hessian of the negative log-likelihood and the prior precision, for a broad class of loss functions. This class includes the Förstner metric for symmetric positive definite matrices, as well as the Kullback--Leibler divergence and the Hellinger distance between the associated distributions. We also propose two fast approximations of the posterior mean and prove their optimality with respect to a weighted Bayes risk under squared-error loss. These approximations are deployed in an offline-online manner, where a more costly but data-independent offline calculation is followed by fast online evaluations. As a result, these approximations are particularly useful when repeated posterior mean evaluations are required for multiple data sets. We demonstrate our theoretical results with several numerical examples, including high-dimensional X-ray tomography and an inverse heat conduction problem. In both of these examples, the intrinsic low-dimensional structure of the inference problem can be exploited while producing results that are essentially indistinguishable from solutions computed in the full space." @default.
• W2963964397 created "2019-07-30" @default.
• W2963964397 creator A5019138446 @default.
• W2963964397 creator A5023616683 @default.
• W2963964397 creator A5038326164 @default.
• W2963964397 creator A5048821900 @default.
• W2963964397 creator A5067931953 @default.
• W2963964397 creator A5071702167 @default.
• W2963964397 date "2015-01-01" @default.
• W2963964397 modified "2023-10-11" @default.
• W2963964397 title "Optimal Low-rank Approximations of Bayesian Linear Inverse Problems" @default.
• W2963964397 cites W1522703524 @default.
• W2963964397 cites W1815314920 @default.
• W2963964397 cites W1837874438 @default.
• W2963964397 cites W1964865031 @default.
• W2963964397 cites W1974511160 @default.
• W2963964397 cites W1976798627 @default.
• W2963964397 cites W1977271127 @default.
• W2963964397 cites W1984000056 @default.
• W2963964397 cites W1985852691 @default.
• W2963964397 cites W1985930395 @default.
• W2963964397 cites W1996375718 @default.
• W2963964397 cites W1999759467 @default.
• W2963964397 cites W2004026774 @default.
• W2963964397 cites W2007500622 @default.
• W2963964397 cites W2009737642 @default.
• W2963964397 cites W2010816198 @default.
• W2963964397 cites W2011718397 @default.
• W2963964397 cites W2015411195 @default.
• W2963964397 cites W2019588162 @default.
• W2963964397 cites W2022217247 @default.
• W2963964397 cites W2050580360 @default.
• W2963964397 cites W2051434435 @default.
• W2963964397 cites W2051669046 @default.
• W2963964397 cites W2058784008 @default.
• W2963964397 cites W2060488365 @default.
• W2963964397 cites W2062243225 @default.
• W2963964397 cites W2068293302 @default.
• W2963964397 cites W2068300001 @default.
• W2963964397 cites W2074686342 @default.
• W2963964397 cites W2074836377 @default.
• W2963964397 cites W2075385817 @default.
• W2963964397 cites W2079559649 @default.
• W2963964397 cites W2082261407 @default.
• W2963964397 cites W2084409811 @default.
• W2963964397 cites W2086351984 @default.
• W2963964397 cites W2091173487 @default.
• W2963964397 cites W2104802194 @default.
• W2963964397 cites W2106222323 @default.
• W2963964397 cites W2117756735 @default.
• W2963964397 cites W2121972410 @default.
• W2963964397 cites W2129728672 @default.
• W2963964397 cites W2138658645 @default.
• W2963964397 cites W2139257007 @default.
• W2963964397 cites W2149498546 @default.
• W2963964397 cites W2165344682 @default.
• W2963964397 cites W2170227272 @default.
• W2963964397 cites W2502336522 @default.
• W2963964397 cites W2962707560 @default.
• W2963964397 cites W4245445339 @default.
• W2963964397 doi "https://doi.org/10.1137/140977308" @default.
• W2963964397 hasPublicationYear "2015" @default.
• W2963964397 type Work @default.
• W2963964397 sameAs 2963964397 @default.
• W2963964397 citedByCount "94" @default.
• W2963964397 countsByYear W29639643972015 @default.
• W2963964397 countsByYear W29639643972016 @default.
• W2963964397 countsByYear W29639643972017 @default.
• W2963964397 countsByYear W29639643972018 @default.
• W2963964397 countsByYear W29639643972019 @default.
• W2963964397 countsByYear W29639643972020 @default.
• W2963964397 countsByYear W29639643972021 @default.
• W2963964397 countsByYear W29639643972022 @default.
• W2963964397 countsByYear W29639643972023 @default.
• W2963964397 crossrefType "journal-article" @default.
• W2963964397 hasAuthorship W2963964397A5019138446 @default.
• W2963964397 hasAuthorship W2963964397A5023616683 @default.
• W2963964397 hasAuthorship W2963964397A5038326164 @default.
• W2963964397 hasAuthorship W2963964397A5048821900 @default.
• W2963964397 hasAuthorship W2963964397A5067931953 @default.
• W2963964397 hasAuthorship W2963964397A5071702167 @default.
• W2963964397 hasBestOaLocation W29639643972 @default.
• W2963964397 hasConcept C105795698 @default.
• W2963964397 hasConcept C107673813 @default.
• W2963964397 hasConcept C11413529 @default.
• W2963964397 hasConcept C134306372 @default.
• W2963964397 hasConcept C135252773 @default.
• W2963964397 hasConcept C138885662 @default.
• W2963964397 hasConcept C162324750 @default.
• W2963964397 hasConcept C171752962 @default.
• W2963964397 hasConcept C176217482 @default.
• W2963964397 hasConcept C177769412 @default.
• W2963964397 hasConcept C185142706 @default.
• W2963964397 hasConcept C203616005 @default.
• W2963964397 hasConcept C207390915 @default.
• W2963964397 hasConcept C21547014 @default.
• W2963964397 hasConcept C28826006 @default.
• W2963964397 hasConcept C33923547 @default.

• next