FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197052110> ?p ?o ?g. }

• W3197052110 abstract "Multivariate, heteroscedastic errors complicate statistical inference in many large-scale denoising problems. Empirical Bayes is attractive in such settings, but standard parametric approaches rest on assumptions about the form of the prior distribution which can be hard to justify and which introduce unnecessary tuning parameters. We extend the nonparametric maximum likelihood estimator (NPMLE) for Gaussian location mixture densities to allow for multivariate, heteroscedastic errors. NPMLEs estimate an arbitrary prior by solving an infinite-dimensional, convex optimization problem; we show that this convex optimization problem can be tractably approximated by a finite-dimensional version. We introduce a dual mixture density whose modes contain the atoms of every NPMLE, and we leverage the dual both to show non-uniqueness in multivariate settings as well as to construct explicit bounds on the support of the NPMLE. The empirical Bayes posterior means based on an NPMLE have low regret, meaning they closely target the oracle posterior means one would compute with the true prior in hand. We prove an oracle inequality implying that the empirical Bayes estimator performs at nearly the optimal level (up to logarithmic factors) for denoising without prior knowledge. We provide finite-sample bounds on the average Hellinger accuracy of an NPMLE for estimating the marginal densities of the observations. We also demonstrate the adaptive and nearly-optimal properties of NPMLEs for deconvolution. We apply the method to two astronomy datasets, constructing a fully data-driven color-magnitude diagram of 1.4 million stars in the Milky Way and investigating the distribution of chemical abundance ratios for 27 thousand stars in the red clump." @default.
• W3197052110 created "2021-09-13" @default.
• W3197052110 creator A5002576969 @default.
• W3197052110 creator A5033475436 @default.
• W3197052110 creator A5084461752 @default.
• W3197052110 date "2021-09-08" @default.
• W3197052110 modified "2023-10-06" @default.
• W3197052110 title "Multivariate, Heteroscedastic Empirical Bayes via Nonparametric Maximum Likelihood" @default.
• W3197052110 cites W1510659740 @default.
• W3197052110 cites W1551986150 @default.
• W3197052110 cites W1560153690 @default.
• W3197052110 cites W1584444527 @default.
• W3197052110 cites W1590636096 @default.
• W3197052110 cites W1965342088 @default.
• W3197052110 cites W1967967135 @default.
• W3197052110 cites W1973255503 @default.
• W3197052110 cites W1981367467 @default.
• W3197052110 cites W1983607152 @default.
• W3197052110 cites W1997817781 @default.
• W3197052110 cites W2000812844 @default.
• W3197052110 cites W2007525292 @default.
• W3197052110 cites W2015241796 @default.
• W3197052110 cites W2015810892 @default.
• W3197052110 cites W2031775363 @default.
• W3197052110 cites W2035220283 @default.
• W3197052110 cites W2045241733 @default.
• W3197052110 cites W2049633694 @default.
• W3197052110 cites W2053201683 @default.
• W3197052110 cites W2057548514 @default.
• W3197052110 cites W2061461017 @default.
• W3197052110 cites W2061868886 @default.
• W3197052110 cites W2072857774 @default.
• W3197052110 cites W2084676098 @default.
• W3197052110 cites W2088349114 @default.
• W3197052110 cites W2091982107 @default.
• W3197052110 cites W2129297925 @default.
• W3197052110 cites W2168036472 @default.
• W3197052110 cites W2170664774 @default.
• W3197052110 cites W2287818844 @default.
• W3197052110 cites W2292172610 @default.
• W3197052110 cites W2461429657 @default.
• W3197052110 cites W249661971 @default.
• W3197052110 cites W2510002191 @default.
• W3197052110 cites W2591785264 @default.
• W3197052110 cites W2744494588 @default.
• W3197052110 cites W2790225006 @default.
• W3197052110 cites W2891574298 @default.
• W3197052110 cites W2949380767 @default.
• W3197052110 cites W2963527517 @default.
• W3197052110 cites W2964344918 @default.
• W3197052110 cites W2982157847 @default.
• W3197052110 cites W2991249605 @default.
• W3197052110 cites W3002620968 @default.
• W3197052110 cites W3006632197 @default.
• W3197052110 cites W3007602320 @default.
• W3197052110 cites W3029017346 @default.
• W3197052110 cites W3034586437 @default.
• W3197052110 cites W3049253527 @default.
• W3197052110 cites W3059391138 @default.
• W3197052110 cites W3098283120 @default.
• W3197052110 cites W3098484589 @default.
• W3197052110 cites W3100196865 @default.
• W3197052110 cites W3102771269 @default.
• W3197052110 cites W3104787397 @default.
• W3197052110 cites W3105658231 @default.
• W3197052110 cites W3105981035 @default.
• W3197052110 cites W3106239789 @default.
• W3197052110 cites W3110538323 @default.
• W3197052110 cites W3132131801 @default.
• W3197052110 cites W3153718095 @default.
• W3197052110 cites W779680562 @default.
• W3197052110 cites W3106246664 @default.
• W3197052110 doi "https://doi.org/10.48550/arxiv.2109.03466" @default.
• W3197052110 hasPublicationYear "2021" @default.
• W3197052110 type Work @default.
• W3197052110 sameAs 3197052110 @default.
• W3197052110 citedByCount "0" @default.
• W3197052110 crossrefType "posted-content" @default.
• W3197052110 hasAuthorship W3197052110A5002576969 @default.
• W3197052110 hasAuthorship W3197052110A5033475436 @default.
• W3197052110 hasAuthorship W3197052110A5084461752 @default.
• W3197052110 hasBestOaLocation W31970521101 @default.
• W3197052110 hasConcept C101104100 @default.
• W3197052110 hasConcept C105795698 @default.
• W3197052110 hasConcept C107673813 @default.
• W3197052110 hasConcept C126255220 @default.
• W3197052110 hasConcept C177769412 @default.
• W3197052110 hasConcept C185429906 @default.
• W3197052110 hasConcept C207201462 @default.
• W3197052110 hasConcept C28826006 @default.
• W3197052110 hasConcept C33923547 @default.
• W3197052110 hasConceptScore W3197052110C101104100 @default.
• W3197052110 hasConceptScore W3197052110C105795698 @default.
• W3197052110 hasConceptScore W3197052110C107673813 @default.
• W3197052110 hasConceptScore W3197052110C126255220 @default.
• W3197052110 hasConceptScore W3197052110C177769412 @default.
• W3197052110 hasConceptScore W3197052110C185429906 @default.
• W3197052110 hasConceptScore W3197052110C207201462 @default.
• W3197052110 hasConceptScore W3197052110C28826006 @default.
• W3197052110 hasConceptScore W3197052110C33923547 @default.

• next