Matches in SemOpenAlex for { <https://semopenalex.org/work/W2952362964> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W2952362964 endingPage "1719" @default.
- W2952362964 startingPage "1709" @default.
- W2952362964 abstract "Latent Dirichlet Allocation (LDA) is a topic model widely used in natural language processing and machine learning. Most approaches to training the model rely on iterative algorithms, which makes it difficult to run LDA on big corpora that are best analyzed in parallel and distributed computational environments. Indeed, current approaches to parallel inference either don't converge to the correct posterior or require storage of large dense matrices in memory. We present a novel sampler that overcomes both problems, and we show that this sampler is faster, both empirically and theoretically, than previous Gibbs samplers for LDA. We do so by employing a novel P'olya-urn-based approximation in the sparse partially collapsed sampler for LDA. We prove that the approximation error vanishes with data size, making our algorithm asymptotically exact, a property of importance for large-scale topic models. In addition, we show, via an explicit example, that - contrary to popular belief in the topic modeling literature - partially collapsed samplers can be more efficient than fully collapsed samplers. We conclude by comparing the performance of our algorithm with that of other approaches on well-known corpora." @default.
- W2952362964 created "2019-06-27" @default.
- W2952362964 creator A5001965218 @default.
- W2952362964 creator A5025191639 @default.
- W2952362964 creator A5037482578 @default.
- W2952362964 creator A5049837768 @default.
- W2952362964 date "2019-07-01" @default.
- W2952362964 modified "2023-10-09" @default.
- W2952362964 title "Pólya Urn Latent Dirichlet Allocation: A Doubly Sparse Massively Parallel Sampler" @default.
- W2952362964 cites W1460189015 @default.
- W2952362964 cites W1515461433 @default.
- W2952362964 cites W1529840045 @default.
- W2952362964 cites W1609010894 @default.
- W2952362964 cites W1889026943 @default.
- W2952362964 cites W2001082470 @default.
- W2952362964 cites W2052261215 @default.
- W2952362964 cites W2063998852 @default.
- W2952362964 cites W2065221212 @default.
- W2952362964 cites W2089484716 @default.
- W2952362964 cites W2090078891 @default.
- W2952362964 cites W2144100511 @default.
- W2952362964 cites W2150731624 @default.
- W2952362964 cites W2487122920 @default.
- W2952362964 cites W2529865518 @default.
- W2952362964 cites W2533991995 @default.
- W2952362964 cites W2963491860 @default.
- W2952362964 doi "https://doi.org/10.1109/tpami.2018.2832641" @default.
- W2952362964 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/29994329" @default.
- W2952362964 hasPublicationYear "2019" @default.
- W2952362964 type Work @default.
- W2952362964 sameAs 2952362964 @default.
- W2952362964 citedByCount "10" @default.
- W2952362964 countsByYear W29523629642017 @default.
- W2952362964 countsByYear W29523629642019 @default.
- W2952362964 countsByYear W29523629642020 @default.
- W2952362964 countsByYear W29523629642022 @default.
- W2952362964 countsByYear W29523629642023 @default.
- W2952362964 crossrefType "journal-article" @default.
- W2952362964 hasAuthorship W2952362964A5001965218 @default.
- W2952362964 hasAuthorship W2952362964A5025191639 @default.
- W2952362964 hasAuthorship W2952362964A5037482578 @default.
- W2952362964 hasAuthorship W2952362964A5049837768 @default.
- W2952362964 hasBestOaLocation W29523629642 @default.
- W2952362964 hasConcept C107673813 @default.
- W2952362964 hasConcept C11413529 @default.
- W2952362964 hasConcept C141318989 @default.
- W2952362964 hasConcept C154945302 @default.
- W2952362964 hasConcept C158424031 @default.
- W2952362964 hasConcept C171686336 @default.
- W2952362964 hasConcept C173608175 @default.
- W2952362964 hasConcept C190475519 @default.
- W2952362964 hasConcept C2776214188 @default.
- W2952362964 hasConcept C2777472644 @default.
- W2952362964 hasConcept C41008148 @default.
- W2952362964 hasConcept C500882744 @default.
- W2952362964 hasConcept C80444323 @default.
- W2952362964 hasConceptScore W2952362964C107673813 @default.
- W2952362964 hasConceptScore W2952362964C11413529 @default.
- W2952362964 hasConceptScore W2952362964C141318989 @default.
- W2952362964 hasConceptScore W2952362964C154945302 @default.
- W2952362964 hasConceptScore W2952362964C158424031 @default.
- W2952362964 hasConceptScore W2952362964C171686336 @default.
- W2952362964 hasConceptScore W2952362964C173608175 @default.
- W2952362964 hasConceptScore W2952362964C190475519 @default.
- W2952362964 hasConceptScore W2952362964C2776214188 @default.
- W2952362964 hasConceptScore W2952362964C2777472644 @default.
- W2952362964 hasConceptScore W2952362964C41008148 @default.
- W2952362964 hasConceptScore W2952362964C500882744 @default.
- W2952362964 hasConceptScore W2952362964C80444323 @default.
- W2952362964 hasFunder F4320320940 @default.
- W2952362964 hasIssue "7" @default.
- W2952362964 hasLocation W29523629641 @default.
- W2952362964 hasLocation W29523629642 @default.
- W2952362964 hasLocation W29523629643 @default.
- W2952362964 hasOpenAccess W2952362964 @default.
- W2952362964 hasPrimaryLocation W29523629641 @default.
- W2952362964 hasRelatedWork W159230833 @default.
- W2952362964 hasRelatedWork W1991255919 @default.
- W2952362964 hasRelatedWork W2148111240 @default.
- W2952362964 hasRelatedWork W2158085718 @default.
- W2952362964 hasRelatedWork W2171278750 @default.
- W2952362964 hasRelatedWork W2279018116 @default.
- W2952362964 hasRelatedWork W2352674739 @default.
- W2952362964 hasRelatedWork W2379152647 @default.
- W2952362964 hasRelatedWork W2388055569 @default.
- W2952362964 hasRelatedWork W2952362964 @default.
- W2952362964 hasVolume "41" @default.
- W2952362964 isParatext "false" @default.
- W2952362964 isRetracted "false" @default.
- W2952362964 magId "2952362964" @default.
- W2952362964 workType "article" @default.