FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W78749393> ?p ?o ?g. }

• W78749393 abstract "Finding the sparsest or minimum e0-norm representation of a signal given a (possibly) overcomplete dictionary of basis vectors is an important problem in many application domains, including neuroelectromagnetic source localization, compressed sensing, sparse component analysis, feature selection, image restoration/compression, and neural coding. Unfortunately, the required optimization is typically NP-hard, and so approximate procedures that succeed with high probability are sought. Nearly all current approaches to this problem, including orthogonal matching pursuit (OMP), basis pursuit (BP) (or the LASSO), and minimum e p quasi-norm methods, can be viewed in Bayesian terms as performing standard MAP estimation using a fixed, sparsity-inducing prior. In contrast, we advocate empirical Bayesian approaches such as sparse Bayesian learning (SBL), which use a parameterized prior to encourage sparsity through a process called evidence maximization. We prove several results about the associated SBL cost function that elucidate its general behavior and provide solid theoretical justification for using it to find maximally sparse representations. Specifically, we show that the global SBL minimum is always achieved at the maximally sparse solution, unlike the BP cost function, while often possessing a more limited constellation of local minima than comparable MAP methods which share this property. We also derive conditions, dependent on the distribution of the nonzero model weights embedded in the optimal representation, such that SBL has no local minima. Finally, we demonstrate how a generalized form of SBL, out of a large class of latent-variable models, uniquely satisfies two minimal performance criteria directly linked to sparsity. These results lead to a deeper understanding of the connections between various Bayesian-inspired strategies and suggest new sparse learning algorithms. Several extensions of SBL are also considered for handling sparse representations that arise in spatio-temporal settings and in the context of covariance component estimation. Here we assume that a small set of common features underly the observed data collected over multiple instances. The theoretical properties of these SBL-based cost functions are examined and evaluated in the context of existing methods. The resulting algorithms display excellent performance on extremely large, ill-posed, and ill-conditioned problems in neuroimaging, suggesting a strong potential for impacting this field and others." @default.
• W78749393 created "2016-06-24" @default.
• W78749393 creator A5001700017 @default.
• W78749393 creator A5085016531 @default.
• W78749393 date "2006-01-01" @default.
• W78749393 modified "2023-09-28" @default.
• W78749393 title "Bayesian methods for finding sparse representations" @default.
• W78749393 cites W1480376833 @default.
• W78749393 cites W1556281046 @default.
• W78749393 cites W1567512734 @default.
• W78749393 cites W1570770495 @default.
• W78749393 cites W1965392255 @default.
• W78749393 cites W1977463519 @default.
• W78749393 cites W1987333243 @default.
• W78749393 cites W1988520084 @default.
• W78749393 cites W1992776590 @default.
• W78749393 cites W2024440951 @default.
• W78749393 cites W2038614136 @default.
• W78749393 cites W2053742104 @default.
• W78749393 cites W2082076175 @default.
• W78749393 cites W2103519107 @default.
• W78749393 cites W2105464873 @default.
• W78749393 cites W2110720597 @default.
• W78749393 cites W2112440119 @default.
• W78749393 cites W2117812871 @default.
• W78749393 cites W2127655853 @default.
• W78749393 cites W2143421693 @default.
• W78749393 cites W2145889472 @default.
• W78749393 cites W2147656689 @default.
• W78749393 cites W2154332973 @default.
• W78749393 cites W2170792849 @default.
• W78749393 cites W2266946488 @default.
• W78749393 cites W2498840900 @default.
• W78749393 cites W2610857016 @default.
• W78749393 cites W2798909945 @default.
• W78749393 cites W2799061466 @default.
• W78749393 cites W3000459232 @default.
• W78749393 hasPublicationYear "2006" @default.
• W78749393 type Work @default.
• W78749393 sameAs 78749393 @default.
• W78749393 citedByCount "23" @default.
• W78749393 countsByYear W787493932012 @default.
• W78749393 countsByYear W787493932013 @default.
• W78749393 countsByYear W787493932014 @default.
• W78749393 countsByYear W787493932015 @default.
• W78749393 countsByYear W787493932016 @default.
• W78749393 countsByYear W787493932017 @default.
• W78749393 countsByYear W787493932019 @default.
• W78749393 countsByYear W787493932020 @default.
• W78749393 crossrefType "dissertation" @default.
• W78749393 hasAuthorship W78749393A5001700017 @default.
• W78749393 hasAuthorship W78749393A5085016531 @default.
• W78749393 hasConcept C107673813 @default.
• W78749393 hasConcept C11413529 @default.
• W78749393 hasConcept C124066611 @default.
• W78749393 hasConcept C124851039 @default.
• W78749393 hasConcept C126255220 @default.
• W78749393 hasConcept C134306372 @default.
• W78749393 hasConcept C136764020 @default.
• W78749393 hasConcept C153180895 @default.
• W78749393 hasConcept C154945302 @default.
• W78749393 hasConcept C156872377 @default.
• W78749393 hasConcept C160234255 @default.
• W78749393 hasConcept C17744445 @default.
• W78749393 hasConcept C186633575 @default.
• W78749393 hasConcept C191795146 @default.
• W78749393 hasConcept C199539241 @default.
• W78749393 hasConcept C33923547 @default.
• W78749393 hasConcept C37616216 @default.
• W78749393 hasConcept C41008148 @default.
• W78749393 hasConcept C5917680 @default.
• W78749393 hasConcept C77637269 @default.
• W78749393 hasConceptScore W78749393C107673813 @default.
• W78749393 hasConceptScore W78749393C11413529 @default.
• W78749393 hasConceptScore W78749393C124066611 @default.
• W78749393 hasConceptScore W78749393C124851039 @default.
• W78749393 hasConceptScore W78749393C126255220 @default.
• W78749393 hasConceptScore W78749393C134306372 @default.
• W78749393 hasConceptScore W78749393C136764020 @default.
• W78749393 hasConceptScore W78749393C153180895 @default.
• W78749393 hasConceptScore W78749393C154945302 @default.
• W78749393 hasConceptScore W78749393C156872377 @default.
• W78749393 hasConceptScore W78749393C160234255 @default.
• W78749393 hasConceptScore W78749393C17744445 @default.
• W78749393 hasConceptScore W78749393C186633575 @default.
• W78749393 hasConceptScore W78749393C191795146 @default.
• W78749393 hasConceptScore W78749393C199539241 @default.
• W78749393 hasConceptScore W78749393C33923547 @default.
• W78749393 hasConceptScore W78749393C37616216 @default.
• W78749393 hasConceptScore W78749393C41008148 @default.
• W78749393 hasConceptScore W78749393C5917680 @default.
• W78749393 hasConceptScore W78749393C77637269 @default.
• W78749393 hasOpenAccess W78749393 @default.
• W78749393 hasRelatedWork W1633751774 @default.
• W78749393 hasRelatedWork W1648445109 @default.
• W78749393 hasRelatedWork W1663973292 @default.
• W78749393 hasRelatedWork W1980405727 @default.
• W78749393 hasRelatedWork W1986931325 @default.
• W78749393 hasRelatedWork W2071284784 @default.
• W78749393 hasRelatedWork W2076776015 @default.

• next