Matches in SemOpenAlex for { <https://semopenalex.org/work/W3120877973> ?p ?o ?g. }
- W3120877973 abstract "The likelihood-informed subspace (LIS) method offers a viable route to reducing the dimensionality of high-dimensional probability distributions arising in Bayesian inference. LIS identifies an intrinsic low-dimensional linear subspace where the target distribution differs the most from some tractable reference distribution. Such a subspace can be identified using the leading eigenvectors of a Gram matrix of the gradient of the log-likelihood function. Then, the original high-dimensional target distribution is approximated through various forms of marginalization of the likelihood function, in which the approximated likelihood only has support on the intrinsic low-dimensional subspace. This approximation enables the design of inference algorithms that can scale sub-linearly with the apparent dimensionality of the problem. Intuitively, the accuracy of the approximation, and hence the performance of the inference algorithms, are influenced by three factors—the dimension truncation error in identifying the subspace, Monte Carlo error in estimating the Gram matrices, and Monte Carlo error in constructing marginalizations. This work establishes a unified framework to analyze each of these three factors and their interplay. Under mild technical assumptions, we establish error bounds for a range of existing dimension reduction techniques based on the principle of LIS. Our error bounds also provide useful insights into the accuracy of these methods. In addition, we analyze the integration of LIS with sampling methods such as Markov Chain Monte Carlo (MCMC) and sequential Monte Carlo (SMC). We also demonstrate the applicability of our analysis on a linear inverse problem with Gaussian prior, which shows that all the estimates can be dimension-independent if the prior covariance is a trace-class operator. Finally, we demonstrate various aspects of our theoretical claims on two nonlinear inverse problems." @default.
- W3120877973 created "2021-01-18" @default.
- W3120877973 creator A5067931953 @default.
- W3120877973 creator A5084545184 @default.
- W3120877973 date "2022-11-01" @default.
- W3120877973 modified "2023-09-26" @default.
- W3120877973 title "A unified performance analysis of likelihood-informed subspace methods" @default.
- W3120877973 cites W1522264804 @default.
- W3120877973 cites W1564223560 @default.
- W3120877973 cites W1996672492 @default.
- W3120877973 cites W1999174665 @default.
- W3120877973 cites W2015846973 @default.
- W3120877973 cites W2054739748 @default.
- W3120877973 cites W2067708435 @default.
- W3120877973 cites W2074836377 @default.
- W3120877973 cites W2082261407 @default.
- W3120877973 cites W2086351984 @default.
- W3120877973 cites W2088911135 @default.
- W3120877973 cites W2091860746 @default.
- W3120877973 cites W2099052275 @default.
- W3120877973 cites W2104067967 @default.
- W3120877973 cites W2129728672 @default.
- W3120877973 cites W2130401121 @default.
- W3120877973 cites W2142737094 @default.
- W3120877973 cites W2143220485 @default.
- W3120877973 cites W2149498546 @default.
- W3120877973 cites W2152657433 @default.
- W3120877973 cites W2152869421 @default.
- W3120877973 cites W2170164530 @default.
- W3120877973 cites W2327680971 @default.
- W3120877973 cites W2418704535 @default.
- W3120877973 cites W2462335633 @default.
- W3120877973 cites W2575742565 @default.
- W3120877973 cites W2603909564 @default.
- W3120877973 cites W2962707560 @default.
- W3120877973 cites W2962811996 @default.
- W3120877973 cites W2962841346 @default.
- W3120877973 cites W2963371806 @default.
- W3120877973 cites W2963641406 @default.
- W3120877973 cites W2963964397 @default.
- W3120877973 cites W2964030972 @default.
- W3120877973 cites W2965797527 @default.
- W3120877973 cites W3000092734 @default.
- W3120877973 cites W3036554994 @default.
- W3120877973 cites W3042743256 @default.
- W3120877973 cites W3098713583 @default.
- W3120877973 cites W3100071005 @default.
- W3120877973 cites W3101449523 @default.
- W3120877973 cites W3104291870 @default.
- W3120877973 cites W3104799488 @default.
- W3120877973 cites W3120877973 @default.
- W3120877973 cites W3134335398 @default.
- W3120877973 cites W3163731508 @default.
- W3120877973 cites W3164056317 @default.
- W3120877973 cites W4233367026 @default.
- W3120877973 cites W4234061158 @default.
- W3120877973 cites W4234524531 @default.
- W3120877973 cites W4246171085 @default.
- W3120877973 cites W4254476808 @default.
- W3120877973 cites W4292156489 @default.
- W3120877973 doi "https://doi.org/10.3150/21-bej1437" @default.
- W3120877973 hasPublicationYear "2022" @default.
- W3120877973 type Work @default.
- W3120877973 sameAs 3120877973 @default.
- W3120877973 citedByCount "5" @default.
- W3120877973 countsByYear W31208779732022 @default.
- W3120877973 crossrefType "journal-article" @default.
- W3120877973 hasAuthorship W3120877973A5067931953 @default.
- W3120877973 hasAuthorship W3120877973A5084545184 @default.
- W3120877973 hasBestOaLocation W31208779732 @default.
- W3120877973 hasConcept C105795698 @default.
- W3120877973 hasConcept C107673813 @default.
- W3120877973 hasConcept C111030470 @default.
- W3120877973 hasConcept C111350023 @default.
- W3120877973 hasConcept C11413529 @default.
- W3120877973 hasConcept C126255220 @default.
- W3120877973 hasConcept C134306372 @default.
- W3120877973 hasConcept C154945302 @default.
- W3120877973 hasConcept C160234255 @default.
- W3120877973 hasConcept C167928553 @default.
- W3120877973 hasConcept C19499675 @default.
- W3120877973 hasConcept C2776214188 @default.
- W3120877973 hasConcept C28826006 @default.
- W3120877973 hasConcept C32834561 @default.
- W3120877973 hasConcept C33923547 @default.
- W3120877973 hasConcept C41008148 @default.
- W3120877973 hasConcept C70518039 @default.
- W3120877973 hasConcept C89106044 @default.
- W3120877973 hasConceptScore W3120877973C105795698 @default.
- W3120877973 hasConceptScore W3120877973C107673813 @default.
- W3120877973 hasConceptScore W3120877973C111030470 @default.
- W3120877973 hasConceptScore W3120877973C111350023 @default.
- W3120877973 hasConceptScore W3120877973C11413529 @default.
- W3120877973 hasConceptScore W3120877973C126255220 @default.
- W3120877973 hasConceptScore W3120877973C134306372 @default.
- W3120877973 hasConceptScore W3120877973C154945302 @default.
- W3120877973 hasConceptScore W3120877973C160234255 @default.
- W3120877973 hasConceptScore W3120877973C167928553 @default.
- W3120877973 hasConceptScore W3120877973C19499675 @default.
- W3120877973 hasConceptScore W3120877973C2776214188 @default.