Matches in SemOpenAlex for { <https://semopenalex.org/work/W2351252181> ?p ?o ?g. }
- W2351252181 abstract "We introduce polyglot language models, recurrent neural network models trained to predict symbol sequences in many different languages using shared representations of symbols and conditioning on typological information about the language to be predicted. We apply these to the problem of modeling phone sequences---a domain in which universal symbol inventories and cross-linguistically shared feature representations are a natural fit. Intrinsic evaluation on held-out perplexity, qualitative analysis of the learned representations, and extrinsic evaluation in two downstream applications that make use of phonetic features show (i) that polyglot models better generalize to held-out data than comparable monolingual models and (ii) that polyglot phonetic feature representations are of higher quality than those learned monolingually." @default.
- W2351252181 created "2016-06-24" @default.
- W2351252181 creator A5005513786 @default.
- W2351252181 creator A5008257790 @default.
- W2351252181 creator A5014855689 @default.
- W2351252181 creator A5042037634 @default.
- W2351252181 creator A5045031910 @default.
- W2351252181 creator A5046364646 @default.
- W2351252181 creator A5054371148 @default.
- W2351252181 creator A5061131385 @default.
- W2351252181 creator A5062910836 @default.
- W2351252181 date "2016-05-12" @default.
- W2351252181 modified "2023-09-27" @default.
- W2351252181 title "Polyglot Neural Language Models: A Case Study in Cross-Lingual Phonetic Representation Learning" @default.
- W2351252181 cites W1522301498 @default.
- W2351252181 cites W1527575280 @default.
- W2351252181 cites W160224403 @default.
- W2351252181 cites W1605308203 @default.
- W2351252181 cites W1783473872 @default.
- W2351252181 cites W1810943226 @default.
- W2351252181 cites W202879582 @default.
- W2351252181 cites W2064675550 @default.
- W2351252181 cites W2080213370 @default.
- W2351252181 cites W2095650036 @default.
- W2351252181 cites W2144404214 @default.
- W2351252181 cites W2157807817 @default.
- W2351252181 cites W2158139315 @default.
- W2351252181 cites W2167949757 @default.
- W2351252181 cites W2171361956 @default.
- W2351252181 cites W2235125105 @default.
- W2351252181 cites W2250192802 @default.
- W2351252181 cites W2250414191 @default.
- W2351252181 cites W2251066368 @default.
- W2351252181 cites W2251204185 @default.
- W2351252181 cites W2251805006 @default.
- W2351252181 cites W2251874715 @default.
- W2351252181 cites W2251939518 @default.
- W2351252181 cites W2252095989 @default.
- W2351252181 cites W2295584157 @default.
- W2351252181 cites W2396366106 @default.
- W2351252181 cites W2402268235 @default.
- W2351252181 cites W2404169761 @default.
- W2351252181 cites W2572569512 @default.
- W2351252181 cites W2949402715 @default.
- W2351252181 cites W2949563612 @default.
- W2351252181 cites W2949952998 @default.
- W2351252181 cites W2964084097 @default.
- W2351252181 cites W3144357634 @default.
- W2351252181 cites W342285082 @default.
- W2351252181 cites W61749939 @default.
- W2351252181 hasPublicationYear "2016" @default.
- W2351252181 type Work @default.
- W2351252181 sameAs 2351252181 @default.
- W2351252181 citedByCount "33" @default.
- W2351252181 countsByYear W23512521812016 @default.
- W2351252181 countsByYear W23512521812017 @default.
- W2351252181 countsByYear W23512521812018 @default.
- W2351252181 countsByYear W23512521812019 @default.
- W2351252181 countsByYear W23512521812020 @default.
- W2351252181 countsByYear W23512521812021 @default.
- W2351252181 crossrefType "posted-content" @default.
- W2351252181 hasAuthorship W2351252181A5005513786 @default.
- W2351252181 hasAuthorship W2351252181A5008257790 @default.
- W2351252181 hasAuthorship W2351252181A5014855689 @default.
- W2351252181 hasAuthorship W2351252181A5042037634 @default.
- W2351252181 hasAuthorship W2351252181A5045031910 @default.
- W2351252181 hasAuthorship W2351252181A5046364646 @default.
- W2351252181 hasAuthorship W2351252181A5054371148 @default.
- W2351252181 hasAuthorship W2351252181A5061131385 @default.
- W2351252181 hasAuthorship W2351252181A5062910836 @default.
- W2351252181 hasConcept C100279451 @default.
- W2351252181 hasConcept C121332964 @default.
- W2351252181 hasConcept C134400042 @default.
- W2351252181 hasConcept C137293760 @default.
- W2351252181 hasConcept C138885662 @default.
- W2351252181 hasConcept C154945302 @default.
- W2351252181 hasConcept C17744445 @default.
- W2351252181 hasConcept C199360897 @default.
- W2351252181 hasConcept C199539241 @default.
- W2351252181 hasConcept C204321447 @default.
- W2351252181 hasConcept C22019652 @default.
- W2351252181 hasConcept C2776359362 @default.
- W2351252181 hasConcept C2776401178 @default.
- W2351252181 hasConcept C2776779415 @default.
- W2351252181 hasConcept C2778707766 @default.
- W2351252181 hasConcept C2780239667 @default.
- W2351252181 hasConcept C28490314 @default.
- W2351252181 hasConcept C30080830 @default.
- W2351252181 hasConcept C41008148 @default.
- W2351252181 hasConcept C41895202 @default.
- W2351252181 hasConcept C50644808 @default.
- W2351252181 hasConcept C59404180 @default.
- W2351252181 hasConcept C62520636 @default.
- W2351252181 hasConcept C94625758 @default.
- W2351252181 hasConceptScore W2351252181C100279451 @default.
- W2351252181 hasConceptScore W2351252181C121332964 @default.
- W2351252181 hasConceptScore W2351252181C134400042 @default.
- W2351252181 hasConceptScore W2351252181C137293760 @default.
- W2351252181 hasConceptScore W2351252181C138885662 @default.
- W2351252181 hasConceptScore W2351252181C154945302 @default.