FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2964308280> ?p ?o ?g. }

• W2964308280 endingPage "1773" @default.
• W2964308280 startingPage "1765" @default.
• W2964308280 abstract "We develop a fully discriminative learning approach for supervised Latent Dirichlet Allocation (LDA) model using Back Propagation (i.e., BP-sLDA), which maximizes the posterior probability of the prediction variable given the input document. Different from traditional variational learning or Gibbs sampling approaches, the proposed learning method applies (i) the mirror descent algorithm for maximum a posterior inference and (ii) back propagation over a deep architecture together with stochastic gradient/mirror descent for model parameter estimation, leading to scalable and end-to-end discriminative learning of the model. As a byproduct, we also apply this technique to develop a new learning method for the traditional unsupervised LDA model (i.e., BP-LDA). Experimental results on three real-world regression and classification tasks show that the proposed methods significantly outperform the previous supervised topic models, neural networks, and is on par with deep neural networks." @default.
• W2964308280 created "2019-07-30" @default.
• W2964308280 creator A5017295309 @default.
• W2964308280 creator A5035017068 @default.
• W2964308280 creator A5043356063 @default.
• W2964308280 creator A5047233371 @default.
• W2964308280 creator A5048159038 @default.
• W2964308280 creator A5060313515 @default.
• W2964308280 creator A5081758689 @default.
• W2964308280 creator A5088339142 @default.
• W2964308280 date "2015-12-07" @default.
• W2964308280 modified "2023-09-24" @default.
• W2964308280 title "End-to-end learning of LDA by mirror-descent back propagation over a deep architecture" @default.
• W2964308280 cites W1505731132 @default.
• W2964308280 cites W1831449718 @default.
• W2964308280 cites W1880262756 @default.
• W2964308280 cites W2001082470 @default.
• W2964308280 cites W2016384870 @default.
• W2964308280 cites W2098062695 @default.
• W2964308280 cites W2100931213 @default.
• W2964308280 cites W2103587173 @default.
• W2964308280 cites W2105231718 @default.
• W2964308280 cites W2106732647 @default.
• W2964308280 cites W2110558001 @default.
• W2964308280 cites W2122683976 @default.
• W2964308280 cites W2127941101 @default.
• W2964308280 cites W2133568543 @default.
• W2964308280 cites W2136189984 @default.
• W2964308280 cites W2136891251 @default.
• W2964308280 cites W2144100511 @default.
• W2964308280 cites W2146502635 @default.
• W2964308280 cites W2160815625 @default.
• W2964308280 cites W2161353674 @default.
• W2964308280 cites W2163302275 @default.
• W2964308280 cites W2186629860 @default.
• W2964308280 cites W2231077521 @default.
• W2964308280 cites W2469994344 @default.
• W2964308280 hasPublicationYear "2015" @default.
• W2964308280 type Work @default.
• W2964308280 sameAs 2964308280 @default.
• W2964308280 citedByCount "8" @default.
• W2964308280 countsByYear W29643082802016 @default.
• W2964308280 countsByYear W29643082802018 @default.
• W2964308280 countsByYear W29643082802019 @default.
• W2964308280 countsByYear W29643082802021 @default.
• W2964308280 crossrefType "proceedings-article" @default.
• W2964308280 hasAuthorship W2964308280A5017295309 @default.
• W2964308280 hasAuthorship W2964308280A5035017068 @default.
• W2964308280 hasAuthorship W2964308280A5043356063 @default.
• W2964308280 hasAuthorship W2964308280A5047233371 @default.
• W2964308280 hasAuthorship W2964308280A5048159038 @default.
• W2964308280 hasAuthorship W2964308280A5060313515 @default.
• W2964308280 hasAuthorship W2964308280A5081758689 @default.
• W2964308280 hasAuthorship W2964308280A5088339142 @default.
• W2964308280 hasConcept C107673813 @default.
• W2964308280 hasConcept C108583219 @default.
• W2964308280 hasConcept C119857082 @default.
• W2964308280 hasConcept C136389625 @default.
• W2964308280 hasConcept C153180895 @default.
• W2964308280 hasConcept C153258448 @default.
• W2964308280 hasConcept C154945302 @default.
• W2964308280 hasConcept C155032097 @default.
• W2964308280 hasConcept C158424031 @default.
• W2964308280 hasConcept C171686336 @default.
• W2964308280 hasConcept C206688291 @default.
• W2964308280 hasConcept C2776214188 @default.
• W2964308280 hasConcept C41008148 @default.
• W2964308280 hasConcept C500882744 @default.
• W2964308280 hasConcept C50644808 @default.
• W2964308280 hasConcept C97931131 @default.
• W2964308280 hasConceptScore W2964308280C107673813 @default.
• W2964308280 hasConceptScore W2964308280C108583219 @default.
• W2964308280 hasConceptScore W2964308280C119857082 @default.
• W2964308280 hasConceptScore W2964308280C136389625 @default.
• W2964308280 hasConceptScore W2964308280C153180895 @default.
• W2964308280 hasConceptScore W2964308280C153258448 @default.
• W2964308280 hasConceptScore W2964308280C154945302 @default.
• W2964308280 hasConceptScore W2964308280C155032097 @default.
• W2964308280 hasConceptScore W2964308280C158424031 @default.
• W2964308280 hasConceptScore W2964308280C171686336 @default.
• W2964308280 hasConceptScore W2964308280C206688291 @default.
• W2964308280 hasConceptScore W2964308280C2776214188 @default.
• W2964308280 hasConceptScore W2964308280C41008148 @default.
• W2964308280 hasConceptScore W2964308280C500882744 @default.
• W2964308280 hasConceptScore W2964308280C50644808 @default.
• W2964308280 hasConceptScore W2964308280C97931131 @default.
• W2964308280 hasLocation W29643082801 @default.
• W2964308280 hasOpenAccess W2964308280 @default.
• W2964308280 hasPrimaryLocation W29643082801 @default.
• W2964308280 hasRelatedWork W1520426560 @default.
• W2964308280 hasRelatedWork W1578488390 @default.
• W2964308280 hasRelatedWork W1880262756 @default.
• W2964308280 hasRelatedWork W2001082470 @default.
• W2964308280 hasRelatedWork W2013533830 @default.
• W2964308280 hasRelatedWork W2098062695 @default.
• W2964308280 hasRelatedWork W2100931213 @default.
• W2964308280 hasRelatedWork W2105093271 @default.
• W2964308280 hasRelatedWork W2133568543 @default.

• next