Matches in SemOpenAlex for { <https://semopenalex.org/work/W3034421838> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W3034421838 abstract "Gaussian process regression networks (GPRN) are powerful Bayesian models for multi-output regression, but their inference is intractable. To address this issue, existing methods use a fully factorized structure (or a mixture of such structures) over all the outputs and latent functions for posterior approximation, which, however, can miss the strong posterior dependencies among the latent variables and hurt the inference quality. In addition, the updates of the variational parameters are inefficient and can be prohibitively expensive for a large number of outputs. To overcome these limitations, we propose a scalable variational inference algorithm for GPRN, which not only captures the abundant posterior dependencies but also is much more efficient for massive outputs. We tensorize the output space and introduce tensor/matrix-normal variational posteriors to capture the posterior correlations and to reduce the parameters. We jointly optimize all the parameters and exploit the inherent Kronecker product structure in the variational model evidence lower bound to accelerate the computation. We demonstrate the advantages of our method in several real-world applications." @default.
- W3034421838 created "2020-06-19" @default.
- W3034421838 creator A5018380170 @default.
- W3034421838 creator A5020806445 @default.
- W3034421838 creator A5024663093 @default.
- W3034421838 creator A5034174266 @default.
- W3034421838 date "2020-07-01" @default.
- W3034421838 modified "2023-09-23" @default.
- W3034421838 title "Scalable Gaussian Process Regression Networks" @default.
- W3034421838 cites W1746819321 @default.
- W3034421838 cites W2076077791 @default.
- W3034421838 cites W2078454401 @default.
- W3034421838 cites W2093402979 @default.
- W3034421838 cites W2119595900 @default.
- W3034421838 cites W2120340025 @default.
- W3034421838 cites W2134894576 @default.
- W3034421838 cites W2140095548 @default.
- W3034421838 cites W2141377530 @default.
- W3034421838 cites W2149775970 @default.
- W3034421838 cites W2158170196 @default.
- W3034421838 cites W2167986580 @default.
- W3034421838 cites W2921400564 @default.
- W3034421838 cites W36503211 @default.
- W3034421838 doi "https://doi.org/10.24963/ijcai.2020/340" @default.
- W3034421838 hasPublicationYear "2020" @default.
- W3034421838 type Work @default.
- W3034421838 sameAs 3034421838 @default.
- W3034421838 citedByCount "2" @default.
- W3034421838 countsByYear W30344218382021 @default.
- W3034421838 crossrefType "proceedings-article" @default.
- W3034421838 hasAuthorship W3034421838A5018380170 @default.
- W3034421838 hasAuthorship W3034421838A5020806445 @default.
- W3034421838 hasAuthorship W3034421838A5024663093 @default.
- W3034421838 hasAuthorship W3034421838A5034174266 @default.
- W3034421838 hasBestOaLocation W30344218381 @default.
- W3034421838 hasConcept C107673813 @default.
- W3034421838 hasConcept C11413529 @default.
- W3034421838 hasConcept C119857082 @default.
- W3034421838 hasConcept C121332964 @default.
- W3034421838 hasConcept C126255220 @default.
- W3034421838 hasConcept C154945302 @default.
- W3034421838 hasConcept C160234255 @default.
- W3034421838 hasConcept C163716315 @default.
- W3034421838 hasConcept C2776214188 @default.
- W3034421838 hasConcept C33923547 @default.
- W3034421838 hasConcept C39482219 @default.
- W3034421838 hasConcept C41008148 @default.
- W3034421838 hasConcept C46030957 @default.
- W3034421838 hasConcept C48044578 @default.
- W3034421838 hasConcept C51167844 @default.
- W3034421838 hasConcept C57830394 @default.
- W3034421838 hasConcept C61326573 @default.
- W3034421838 hasConcept C62520636 @default.
- W3034421838 hasConcept C77088390 @default.
- W3034421838 hasConceptScore W3034421838C107673813 @default.
- W3034421838 hasConceptScore W3034421838C11413529 @default.
- W3034421838 hasConceptScore W3034421838C119857082 @default.
- W3034421838 hasConceptScore W3034421838C121332964 @default.
- W3034421838 hasConceptScore W3034421838C126255220 @default.
- W3034421838 hasConceptScore W3034421838C154945302 @default.
- W3034421838 hasConceptScore W3034421838C160234255 @default.
- W3034421838 hasConceptScore W3034421838C163716315 @default.
- W3034421838 hasConceptScore W3034421838C2776214188 @default.
- W3034421838 hasConceptScore W3034421838C33923547 @default.
- W3034421838 hasConceptScore W3034421838C39482219 @default.
- W3034421838 hasConceptScore W3034421838C41008148 @default.
- W3034421838 hasConceptScore W3034421838C46030957 @default.
- W3034421838 hasConceptScore W3034421838C48044578 @default.
- W3034421838 hasConceptScore W3034421838C51167844 @default.
- W3034421838 hasConceptScore W3034421838C57830394 @default.
- W3034421838 hasConceptScore W3034421838C61326573 @default.
- W3034421838 hasConceptScore W3034421838C62520636 @default.
- W3034421838 hasConceptScore W3034421838C77088390 @default.
- W3034421838 hasLocation W30344218381 @default.
- W3034421838 hasOpenAccess W3034421838 @default.
- W3034421838 hasPrimaryLocation W30344218381 @default.
- W3034421838 hasRelatedWork W10593963 @default.
- W3034421838 hasRelatedWork W10943332 @default.
- W3034421838 hasRelatedWork W11165844 @default.
- W3034421838 hasRelatedWork W12726742 @default.
- W3034421838 hasRelatedWork W1362903 @default.
- W3034421838 hasRelatedWork W14023250 @default.
- W3034421838 hasRelatedWork W1584117 @default.
- W3034421838 hasRelatedWork W6799292 @default.
- W3034421838 hasRelatedWork W9573264 @default.
- W3034421838 hasRelatedWork W9917914 @default.
- W3034421838 isParatext "false" @default.
- W3034421838 isRetracted "false" @default.
- W3034421838 magId "3034421838" @default.
- W3034421838 workType "article" @default.