Matches in SemOpenAlex for { <https://semopenalex.org/work/W2899562840> ?p ?o ?g. }
- W2899562840 abstract "We investigate penalized maximum log-likelihood estimation for exponential family distributions whose natural parameter resides in a reproducing kernel Hilbert space. Key to our approach is a novel technique, doubly dual embedding, that avoids computation of the partition function. This technique also allows the development of a flexible sampling strategy that amortizes the cost of Monte-Carlo sampling in the inference stage. The resulting estimator can be easily generalized to kernel conditional exponential families. We establish a connection between kernel exponential family estimation and MMD-GANs, revealing a new perspective for understanding GANs. Compared to the score matching based estimators, the proposed method improves both memory and time efficiency while enjoying stronger statistical properties, such as fully capturing smoothness in its statistical convergence rate while the score matching estimator appears to saturate. Finally, we show that the proposed estimator empirically outperforms state-of-the-art" @default.
- W2899562840 created "2018-11-16" @default.
- W2899562840 creator A5004160420 @default.
- W2899562840 creator A5010575626 @default.
- W2899562840 creator A5030589527 @default.
- W2899562840 creator A5040269658 @default.
- W2899562840 creator A5071683073 @default.
- W2899562840 creator A5086484914 @default.
- W2899562840 date "2018-11-06" @default.
- W2899562840 modified "2023-10-11" @default.
- W2899562840 title "Kernel Exponential Family Estimation via Doubly Dual Embedding" @default.
- W2899562840 cites W122050270 @default.
- W2899562840 cites W1496467656 @default.
- W2899562840 cites W1503425191 @default.
- W2899562840 cites W1506153731 @default.
- W2899562840 cites W1513873506 @default.
- W2899562840 cites W1551944942 @default.
- W2899562840 cites W1881478816 @default.
- W2899562840 cites W1959608418 @default.
- W2899562840 cites W1985077192 @default.
- W2899562840 cites W1992208280 @default.
- W2899562840 cites W2015953335 @default.
- W2899562840 cites W2020247984 @default.
- W2899562840 cites W2040335964 @default.
- W2899562840 cites W2069663974 @default.
- W2899562840 cites W2099471712 @default.
- W2899562840 cites W2115003579 @default.
- W2899562840 cites W2116064496 @default.
- W2899562840 cites W2123395972 @default.
- W2899562840 cites W2144902422 @default.
- W2899562840 cites W2145763471 @default.
- W2899562840 cites W2150621701 @default.
- W2899562840 cites W2158231134 @default.
- W2899562840 cites W2169269897 @default.
- W2899562840 cites W2587284713 @default.
- W2899562840 cites W2806985155 @default.
- W2899562840 cites W2962695743 @default.
- W2899562840 cites W2962737466 @default.
- W2899562840 cites W2962892300 @default.
- W2899562840 cites W2962897886 @default.
- W2899562840 cites W2962919088 @default.
- W2899562840 cites W2963090522 @default.
- W2899562840 cites W2963470657 @default.
- W2899562840 cites W2963800509 @default.
- W2899562840 cites W2963972841 @default.
- W2899562840 cites W2963993553 @default.
- W2899562840 cites W2964224666 @default.
- W2899562840 cites W342858109 @default.
- W2899562840 cites W651031249 @default.
- W2899562840 hasPublicationYear "2018" @default.
- W2899562840 type Work @default.
- W2899562840 sameAs 2899562840 @default.
- W2899562840 citedByCount "0" @default.
- W2899562840 crossrefType "posted-content" @default.
- W2899562840 hasAuthorship W2899562840A5004160420 @default.
- W2899562840 hasAuthorship W2899562840A5010575626 @default.
- W2899562840 hasAuthorship W2899562840A5030589527 @default.
- W2899562840 hasAuthorship W2899562840A5040269658 @default.
- W2899562840 hasAuthorship W2899562840A5071683073 @default.
- W2899562840 hasAuthorship W2899562840A5086484914 @default.
- W2899562840 hasConcept C105795698 @default.
- W2899562840 hasConcept C11413529 @default.
- W2899562840 hasConcept C118615104 @default.
- W2899562840 hasConcept C122280245 @default.
- W2899562840 hasConcept C12267149 @default.
- W2899562840 hasConcept C126255220 @default.
- W2899562840 hasConcept C134517425 @default.
- W2899562840 hasConcept C154945302 @default.
- W2899562840 hasConcept C185429906 @default.
- W2899562840 hasConcept C195699287 @default.
- W2899562840 hasConcept C28826006 @default.
- W2899562840 hasConcept C33923547 @default.
- W2899562840 hasConcept C41008148 @default.
- W2899562840 hasConcept C55974624 @default.
- W2899562840 hasConcept C74193536 @default.
- W2899562840 hasConceptScore W2899562840C105795698 @default.
- W2899562840 hasConceptScore W2899562840C11413529 @default.
- W2899562840 hasConceptScore W2899562840C118615104 @default.
- W2899562840 hasConceptScore W2899562840C122280245 @default.
- W2899562840 hasConceptScore W2899562840C12267149 @default.
- W2899562840 hasConceptScore W2899562840C126255220 @default.
- W2899562840 hasConceptScore W2899562840C134517425 @default.
- W2899562840 hasConceptScore W2899562840C154945302 @default.
- W2899562840 hasConceptScore W2899562840C185429906 @default.
- W2899562840 hasConceptScore W2899562840C195699287 @default.
- W2899562840 hasConceptScore W2899562840C28826006 @default.
- W2899562840 hasConceptScore W2899562840C33923547 @default.
- W2899562840 hasConceptScore W2899562840C41008148 @default.
- W2899562840 hasConceptScore W2899562840C55974624 @default.
- W2899562840 hasConceptScore W2899562840C74193536 @default.
- W2899562840 hasLocation W28995628401 @default.
- W2899562840 hasOpenAccess W2899562840 @default.
- W2899562840 hasPrimaryLocation W28995628401 @default.
- W2899562840 hasRelatedWork W1546708447 @default.
- W2899562840 hasRelatedWork W2082483588 @default.
- W2899562840 hasRelatedWork W2125972037 @default.
- W2899562840 hasRelatedWork W2130314681 @default.
- W2899562840 hasRelatedWork W2151244157 @default.
- W2899562840 hasRelatedWork W2261846873 @default.
- W2899562840 hasRelatedWork W2345897661 @default.