FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1609010894> ?p ?o ?g. }

• W1609010894 abstract "When building large-scale machine learning (ML) programs, such as massive topic models or deep neural networks with up to trillions of parameters and training examples, one usually assumes that such massive tasks can only be attempted with industrial-sized clusters with thousands of nodes, which are out of reach for most practitioners and academic researchers. We consider this challenge in the context of topic modeling on web-scale corpora, and show that with a modest cluster of as few as 8 machines, we can train a topic model with 1 million topics and a 1-million-word vocabulary (for a total of 1 trillion parameters), on a document collection with 200 billion tokens --- a scale not yet reported even with thousands of machines. Our major contributions include: 1) a new, highly-efficient O(1) Metropolis-Hastings sampling algorithm, whose running cost is (surprisingly) agnostic of model size, and empirically converges nearly an order of magnitude more quickly than current state-of-the-art Gibbs samplers; 2) a model-scheduling scheme to handle the big model challenge, where each worker machine schedules the fetch/use of sub-models as needed, resulting in a frugal use of limited memory capacity and network bandwidth; 3) a differential data-structure for model storage, which uses separate data structures for high- and low-frequency words to allow extremely large models to fit in memory, while maintaining high inference speed. These contributions are built on top of the Petuum open-source distributed ML framework, and we provide experimental evidence showing how this development puts massive data and models within reach on a small cluster, while still enjoying proportional time cost reductions with increasing cluster size." @default.
• W1609010894 created "2016-06-24" @default.
• W1609010894 creator A5009547049 @default.
• W1609010894 creator A5012361506 @default.
• W1609010894 creator A5023132445 @default.
• W1609010894 creator A5036378701 @default.
• W1609010894 creator A5053966682 @default.
• W1609010894 creator A5057783300 @default.
• W1609010894 creator A5068888494 @default.
• W1609010894 creator A5070990160 @default.
• W1609010894 creator A5085382296 @default.
• W1609010894 date "2015-05-18" @default.
• W1609010894 modified "2023-10-01" @default.
• W1609010894 title "LightLDA" @default.
• W1609010894 cites W1866637071 @default.
• W1609010894 cites W2001082470 @default.
• W1609010894 cites W2041517243 @default.
• W1609010894 cites W2049875313 @default.
• W1609010894 cites W2052261215 @default.
• W1609010894 cites W2056760934 @default.
• W1609010894 cites W2058209916 @default.
• W1609010894 cites W2065221212 @default.
• W1609010894 cites W2083842231 @default.
• W1609010894 cites W2087937280 @default.
• W1609010894 cites W2101101940 @default.
• W1609010894 cites W2135194391 @default.
• W1609010894 cites W2136796925 @default.
• W1609010894 cites W2138309709 @default.
• W1609010894 cites W2138996412 @default.
• W1609010894 cites W2150731624 @default.
• W1609010894 cites W3003241580 @default.
• W1609010894 doi "https://doi.org/10.1145/2736277.2741115" @default.
• W1609010894 hasPublicationYear "2015" @default.
• W1609010894 type Work @default.
• W1609010894 sameAs 1609010894 @default.
• W1609010894 citedByCount "138" @default.
• W1609010894 countsByYear W16090108942013 @default.
• W1609010894 countsByYear W16090108942014 @default.
• W1609010894 countsByYear W16090108942015 @default.
• W1609010894 countsByYear W16090108942016 @default.
• W1609010894 countsByYear W16090108942017 @default.
• W1609010894 countsByYear W16090108942018 @default.
• W1609010894 countsByYear W16090108942019 @default.
• W1609010894 countsByYear W16090108942020 @default.
• W1609010894 countsByYear W16090108942021 @default.
• W1609010894 countsByYear W16090108942022 @default.
• W1609010894 countsByYear W16090108942023 @default.
• W1609010894 crossrefType "proceedings-article" @default.
• W1609010894 hasAuthorship W1609010894A5009547049 @default.
• W1609010894 hasAuthorship W1609010894A5012361506 @default.
• W1609010894 hasAuthorship W1609010894A5023132445 @default.
• W1609010894 hasAuthorship W1609010894A5036378701 @default.
• W1609010894 hasAuthorship W1609010894A5053966682 @default.
• W1609010894 hasAuthorship W1609010894A5057783300 @default.
• W1609010894 hasAuthorship W1609010894A5068888494 @default.
• W1609010894 hasAuthorship W1609010894A5070990160 @default.
• W1609010894 hasAuthorship W1609010894A5085382296 @default.
• W1609010894 hasConcept C119857082 @default.
• W1609010894 hasConcept C138885662 @default.
• W1609010894 hasConcept C151730666 @default.
• W1609010894 hasConcept C154945302 @default.
• W1609010894 hasConcept C162324750 @default.
• W1609010894 hasConcept C206729178 @default.
• W1609010894 hasConcept C21547014 @default.
• W1609010894 hasConcept C2776214188 @default.
• W1609010894 hasConcept C2777601683 @default.
• W1609010894 hasConcept C2779343474 @default.
• W1609010894 hasConcept C2984842247 @default.
• W1609010894 hasConcept C41008148 @default.
• W1609010894 hasConcept C41895202 @default.
• W1609010894 hasConcept C50644808 @default.
• W1609010894 hasConcept C86803240 @default.
• W1609010894 hasConceptScore W1609010894C119857082 @default.
• W1609010894 hasConceptScore W1609010894C138885662 @default.
• W1609010894 hasConceptScore W1609010894C151730666 @default.
• W1609010894 hasConceptScore W1609010894C154945302 @default.
• W1609010894 hasConceptScore W1609010894C162324750 @default.
• W1609010894 hasConceptScore W1609010894C206729178 @default.
• W1609010894 hasConceptScore W1609010894C21547014 @default.
• W1609010894 hasConceptScore W1609010894C2776214188 @default.
• W1609010894 hasConceptScore W1609010894C2777601683 @default.
• W1609010894 hasConceptScore W1609010894C2779343474 @default.
• W1609010894 hasConceptScore W1609010894C2984842247 @default.
• W1609010894 hasConceptScore W1609010894C41008148 @default.
• W1609010894 hasConceptScore W1609010894C41895202 @default.
• W1609010894 hasConceptScore W1609010894C50644808 @default.
• W1609010894 hasConceptScore W1609010894C86803240 @default.
• W1609010894 hasLocation W16090108941 @default.
• W1609010894 hasOpenAccess W1609010894 @default.
• W1609010894 hasPrimaryLocation W16090108941 @default.
• W1609010894 hasRelatedWork W2961085424 @default.
• W1609010894 hasRelatedWork W3038776261 @default.
• W1609010894 hasRelatedWork W3046775127 @default.
• W1609010894 hasRelatedWork W3170094116 @default.
• W1609010894 hasRelatedWork W3205684019 @default.
• W1609010894 hasRelatedWork W4285260836 @default.
• W1609010894 hasRelatedWork W4286629047 @default.
• W1609010894 hasRelatedWork W4306321456 @default.
• W1609010894 hasRelatedWork W4306674287 @default.
• W1609010894 hasRelatedWork W4224009465 @default.

• next