Matches in SemOpenAlex for { <https://semopenalex.org/work/W2901350054> ?p ?o ?g. }
- W2901350054 abstract "The problem of posterior inference for individual documents is particularly important in topic models. However, it is often intractable in practice. Many existing methods for posterior inference such as variational Bayes, collapsed variational Bayes and collapsed Gibbs sampling do not have any guarantee on either quality or rate of convergence. The online maximum a posteriori estimation (OPE) algorithm has more attractive properties than other inference approaches. In this paper, we introduced four algorithms to improve OPE (namely, OPE1, OPE2, OPE3, and OPE4) by combining two stochastic bounds. Our new algorithms not only preserve the key advantages of OPE but also can sometimes perform significantly better than OPE. These algorithms were employed to develop new effective methods for learning topic models from massive/streaming text collections. Empirical results show that our approaches were often more efficient than the state-of-theart methods. DOI: 10.32913/rd-ict.vol2.no15.687" @default.
- W2901350054 created "2018-11-29" @default.
- W2901350054 creator A5035216205 @default.
- W2901350054 creator A5054854013 @default.
- W2901350054 creator A5082954989 @default.
- W2901350054 date "2018-08-11" @default.
- W2901350054 modified "2023-09-24" @default.
- W2901350054 title "Some Methods for Posterior Inference in Topic Models" @default.
- W2901350054 cites W1880262756 @default.
- W2901350054 cites W2087648362 @default.
- W2901350054 cites W2098126593 @default.
- W2901350054 cites W2109706083 @default.
- W2901350054 cites W2110558001 @default.
- W2901350054 cites W2111200589 @default.
- W2901350054 cites W2116350082 @default.
- W2901350054 cites W2130339025 @default.
- W2901350054 cites W2151178372 @default.
- W2901350054 cites W2153383412 @default.
- W2901350054 cites W2161353674 @default.
- W2901350054 cites W2174706414 @default.
- W2901350054 cites W2251582277 @default.
- W2901350054 cites W2481528776 @default.
- W2901350054 cites W2949177896 @default.
- W2901350054 cites W2950662148 @default.
- W2901350054 cites W2950770596 @default.
- W2901350054 cites W2963625764 @default.
- W2901350054 cites W3101380508 @default.
- W2901350054 cites W319974279 @default.
- W2901350054 doi "https://doi.org/10.32913/rd-ict.vol2.no15.687" @default.
- W2901350054 hasPublicationYear "2018" @default.
- W2901350054 type Work @default.
- W2901350054 sameAs 2901350054 @default.
- W2901350054 citedByCount "0" @default.
- W2901350054 crossrefType "journal-article" @default.
- W2901350054 hasAuthorship W2901350054A5035216205 @default.
- W2901350054 hasAuthorship W2901350054A5054854013 @default.
- W2901350054 hasAuthorship W2901350054A5082954989 @default.
- W2901350054 hasBestOaLocation W29013500541 @default.
- W2901350054 hasConcept C105795698 @default.
- W2901350054 hasConcept C107673813 @default.
- W2901350054 hasConcept C111472728 @default.
- W2901350054 hasConcept C11413529 @default.
- W2901350054 hasConcept C119857082 @default.
- W2901350054 hasConcept C126255220 @default.
- W2901350054 hasConcept C138885662 @default.
- W2901350054 hasConcept C154945302 @default.
- W2901350054 hasConcept C158424031 @default.
- W2901350054 hasConcept C160234255 @default.
- W2901350054 hasConcept C162324750 @default.
- W2901350054 hasConcept C207201462 @default.
- W2901350054 hasConcept C26517878 @default.
- W2901350054 hasConcept C2776214188 @default.
- W2901350054 hasConcept C2777303404 @default.
- W2901350054 hasConcept C33923547 @default.
- W2901350054 hasConcept C38652104 @default.
- W2901350054 hasConcept C41008148 @default.
- W2901350054 hasConcept C49781872 @default.
- W2901350054 hasConcept C50522688 @default.
- W2901350054 hasConcept C57830394 @default.
- W2901350054 hasConcept C75553542 @default.
- W2901350054 hasConcept C9810830 @default.
- W2901350054 hasConceptScore W2901350054C105795698 @default.
- W2901350054 hasConceptScore W2901350054C107673813 @default.
- W2901350054 hasConceptScore W2901350054C111472728 @default.
- W2901350054 hasConceptScore W2901350054C11413529 @default.
- W2901350054 hasConceptScore W2901350054C119857082 @default.
- W2901350054 hasConceptScore W2901350054C126255220 @default.
- W2901350054 hasConceptScore W2901350054C138885662 @default.
- W2901350054 hasConceptScore W2901350054C154945302 @default.
- W2901350054 hasConceptScore W2901350054C158424031 @default.
- W2901350054 hasConceptScore W2901350054C160234255 @default.
- W2901350054 hasConceptScore W2901350054C162324750 @default.
- W2901350054 hasConceptScore W2901350054C207201462 @default.
- W2901350054 hasConceptScore W2901350054C26517878 @default.
- W2901350054 hasConceptScore W2901350054C2776214188 @default.
- W2901350054 hasConceptScore W2901350054C2777303404 @default.
- W2901350054 hasConceptScore W2901350054C33923547 @default.
- W2901350054 hasConceptScore W2901350054C38652104 @default.
- W2901350054 hasConceptScore W2901350054C41008148 @default.
- W2901350054 hasConceptScore W2901350054C49781872 @default.
- W2901350054 hasConceptScore W2901350054C50522688 @default.
- W2901350054 hasConceptScore W2901350054C57830394 @default.
- W2901350054 hasConceptScore W2901350054C75553542 @default.
- W2901350054 hasConceptScore W2901350054C9810830 @default.
- W2901350054 hasLocation W29013500541 @default.
- W2901350054 hasOpenAccess W2901350054 @default.
- W2901350054 hasPrimaryLocation W29013500541 @default.
- W2901350054 hasRelatedWork W2314324790 @default.
- W2901350054 hasRelatedWork W2335394797 @default.
- W2901350054 hasRelatedWork W2511279186 @default.
- W2901350054 hasRelatedWork W2605000734 @default.
- W2901350054 hasRelatedWork W2901350054 @default.
- W2901350054 hasRelatedWork W2963058055 @default.
- W2901350054 hasRelatedWork W3034669169 @default.
- W2901350054 hasRelatedWork W3161619868 @default.
- W2901350054 hasRelatedWork W4226326567 @default.
- W2901350054 hasRelatedWork W4287870652 @default.
- W2901350054 isParatext "false" @default.
- W2901350054 isRetracted "false" @default.
- W2901350054 magId "2901350054" @default.