FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963177712> ?p ?o ?g. }

• W2963177712 endingPage "670" @default.
• W2963177712 startingPage "650" @default.
• W2963177712 abstract "Maximum-a-posteriori (MAP) estimation is the main Bayesian estimation methodology in imaging sciences, where high dimensionality is often addressed by using Bayesian models that are log-concave and whose posterior mode can be computed efficiently by convex optimization. However, despite its success and wide adoption, MAP estimation is not theoretically well understood yet. In particular, the prevalent view in the community is that MAP estimation is not proper Bayesian estimation in the sense of Bayesian decision theory because it does not minimize a meaningful expected loss function (unlike the minimum mean squared error (MMSE) estimator that minimizes the mean squared loss). This paper addresses this theoretical gap by presenting a general decision-theoretic derivation of MAP estimation in Bayesian models that are log-concave. A main novelty is that our analysis is based on differential geometry and proceeds as follows. First, we use the underlying convex geometry of the Bayesian model to induce a Riemannian geometry on the parameter space. We then use differential geometry to identify the so-called natural or canonical loss function to perform Bayesian point estimation in that Riemannian manifold. For log-concave models, this canonical loss coincides with the Bregman divergence associated with the negative log posterior density. Following on from this, we show that the MAP estimator is the only Bayesian estimator that minimizes the expected canonical loss, and that the posterior mean or MMSE estimator minimizes the dual canonical loss. We then study the question of MAP and MMSE estimation performance in high dimensions. Precisely, we establish a universal bound on the expected canonical error as a function of image dimension, providing new insights on the good empirical performance observed in convex problems. Together, these results provide a new understanding of MAP and MMSE estimation in log-concave settings, and of the multiple beneficial roles that convex geometry plays in imaging problems. Finally, we illustrate this new theory by analyzing the regularization-by-denoising Bayesian models, a class of state-of-the-art imaging models where priors are defined implicitly through image denoising algorithms, and an image denoising model with a wavelet shrinkage prior." @default.
• W2963177712 created "2019-07-30" @default.
• W2963177712 creator A5082169271 @default.
• W2963177712 date "2019-01-01" @default.
• W2963177712 modified "2023-09-25" @default.
• W2963177712 title "Revisiting Maximum-A-Posteriori Estimation in Log-Concave Models" @default.
• W2963177712 cites W1774344329 @default.
• W2963177712 cites W2064991026 @default.
• W2963177712 cites W2065513175 @default.
• W2963177712 cites W2096567125 @default.
• W2963177712 cites W2105536051 @default.
• W2963177712 cites W2105910985 @default.
• W2963177712 cites W2108871462 @default.
• W2963177712 cites W2121952483 @default.
• W2963177712 cites W2138598313 @default.
• W2963177712 cites W2141006018 @default.
• W2963177712 cites W2145096794 @default.
• W2963177712 cites W2158755508 @default.
• W2963177712 cites W2193424816 @default.
• W2963177712 cites W2405803404 @default.
• W2963177712 cites W2566924527 @default.
• W2963177712 cites W2573726823 @default.
• W2963177712 cites W2611328865 @default.
• W2963177712 cites W2612986376 @default.
• W2963177712 cites W2768751982 @default.
• W2963177712 cites W2963746556 @default.
• W2963177712 cites W3100456593 @default.
• W2963177712 cites W3103995022 @default.
• W2963177712 cites W3106359998 @default.
• W2963177712 cites W3121742466 @default.
• W2963177712 cites W4242350695 @default.
• W2963177712 cites W639587122 @default.
• W2963177712 doi "https://doi.org/10.1137/18m1174076" @default.
• W2963177712 hasPublicationYear "2019" @default.
• W2963177712 type Work @default.
• W2963177712 sameAs 2963177712 @default.
• W2963177712 citedByCount "15" @default.
• W2963177712 countsByYear W29631777122019 @default.
• W2963177712 countsByYear W29631777122020 @default.
• W2963177712 countsByYear W29631777122021 @default.
• W2963177712 countsByYear W29631777122022 @default.
• W2963177712 countsByYear W29631777122023 @default.
• W2963177712 crossrefType "journal-article" @default.
• W2963177712 hasAuthorship W2963177712A5082169271 @default.
• W2963177712 hasBestOaLocation W29631777122 @default.
• W2963177712 hasConcept C105795698 @default.
• W2963177712 hasConcept C107673813 @default.
• W2963177712 hasConcept C11413529 @default.
• W2963177712 hasConcept C126255220 @default.
• W2963177712 hasConcept C139945424 @default.
• W2963177712 hasConcept C149073432 @default.
• W2963177712 hasConcept C153180895 @default.
• W2963177712 hasConcept C154945302 @default.
• W2963177712 hasConcept C185429906 @default.
• W2963177712 hasConcept C28826006 @default.
• W2963177712 hasConcept C33923547 @default.
• W2963177712 hasConcept C41008148 @default.
• W2963177712 hasConcept C49781872 @default.
• W2963177712 hasConcept C68022304 @default.
• W2963177712 hasConcept C9810830 @default.
• W2963177712 hasConceptScore W2963177712C105795698 @default.
• W2963177712 hasConceptScore W2963177712C107673813 @default.
• W2963177712 hasConceptScore W2963177712C11413529 @default.
• W2963177712 hasConceptScore W2963177712C126255220 @default.
• W2963177712 hasConceptScore W2963177712C139945424 @default.
• W2963177712 hasConceptScore W2963177712C149073432 @default.
• W2963177712 hasConceptScore W2963177712C153180895 @default.
• W2963177712 hasConceptScore W2963177712C154945302 @default.
• W2963177712 hasConceptScore W2963177712C185429906 @default.
• W2963177712 hasConceptScore W2963177712C28826006 @default.
• W2963177712 hasConceptScore W2963177712C33923547 @default.
• W2963177712 hasConceptScore W2963177712C41008148 @default.
• W2963177712 hasConceptScore W2963177712C49781872 @default.
• W2963177712 hasConceptScore W2963177712C68022304 @default.
• W2963177712 hasConceptScore W2963177712C9810830 @default.
• W2963177712 hasFunder F4320334678 @default.
• W2963177712 hasIssue "1" @default.
• W2963177712 hasLocation W29631777121 @default.
• W2963177712 hasLocation W29631777122 @default.
• W2963177712 hasOpenAccess W2963177712 @default.
• W2963177712 hasPrimaryLocation W29631777121 @default.
• W2963177712 hasRelatedWork W106751956 @default.
• W2963177712 hasRelatedWork W148593058 @default.
• W2963177712 hasRelatedWork W2147265815 @default.
• W2963177712 hasRelatedWork W214727523 @default.
• W2963177712 hasRelatedWork W2158892545 @default.
• W2963177712 hasRelatedWork W2187942274 @default.
• W2963177712 hasRelatedWork W2464015051 @default.
• W2963177712 hasRelatedWork W2521753262 @default.
• W2963177712 hasRelatedWork W2582182730 @default.
• W2963177712 hasRelatedWork W2907746047 @default.
• W2963177712 hasVolume "12" @default.
• W2963177712 isParatext "false" @default.
• W2963177712 isRetracted "false" @default.
• W2963177712 magId "2963177712" @default.
• W2963177712 workType "article" @default.