FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2289096394> ?p ?o ?g. }

• W2289096394 abstract "Author(s): Sedghi, Hanie; Anandkumar, Anima | Abstract: We consider the problem of training input-output recurrent neural networks (RNN) for sequence labeling tasks. We propose a novel spectral approach for learning the network parameters. It is based on decomposition of the cross-moment tensor between the output and a non-linear transformation of the input, based on score functions. We guarantee consistent learning with polynomial sample and computational complexity under transparent conditions such as non-degeneracy of model parameters, polynomial activations for the neurons, and a Markovian evolution of the input sequence. We also extend our results to Bidirectional RNN which uses both previous and future information to output the label at each time point, and is employed in many NLP tasks such as POS tagging." @default.
• W2289096394 created "2016-06-24" @default.
• W2289096394 creator A5014498545 @default.
• W2289096394 creator A5045805879 @default.
• W2289096394 date "2016-03-02" @default.
• W2289096394 modified "2023-09-27" @default.
• W2289096394 title "Training Input-Output Recurrent Neural Networks through Spectral Methods" @default.
• W2289096394 cites W1505878979 @default.
• W2289096394 cites W1574901103 @default.
• W2289096394 cites W1581152950 @default.
• W2289096394 cites W1583477766 @default.
• W2289096394 cites W1598796236 @default.
• W2289096394 cites W1811734137 @default.
• W2289096394 cites W1839868949 @default.
• W2289096394 cites W1955967056 @default.
• W2289096394 cites W2034797247 @default.
• W2289096394 cites W2098650530 @default.
• W2289096394 cites W2105724942 @default.
• W2289096394 cites W2116059156 @default.
• W2289096394 cites W2131774270 @default.
• W2289096394 cites W2162636131 @default.
• W2289096394 cites W2164039623 @default.
• W2289096394 cites W2191540403 @default.
• W2289096394 cites W2250861254 @default.
• W2289096394 cites W2251937901 @default.
• W2289096394 cites W2294384375 @default.
• W2289096394 cites W2407775036 @default.
• W2289096394 cites W2949531434 @default.
• W2289096394 cites W2950289646 @default.
• W2289096394 cites W2950297649 @default.
• W2289096394 cites W2953337630 @default.
• W2289096394 cites W339865234 @default.
• W2289096394 hasPublicationYear "2016" @default.
• W2289096394 type Work @default.
• W2289096394 sameAs 2289096394 @default.
• W2289096394 citedByCount "10" @default.
• W2289096394 countsByYear W22890963942016 @default.
• W2289096394 countsByYear W22890963942017 @default.
• W2289096394 countsByYear W22890963942018 @default.
• W2289096394 crossrefType "posted-content" @default.
• W2289096394 hasAuthorship W2289096394A5014498545 @default.
• W2289096394 hasAuthorship W2289096394A5045805879 @default.
• W2289096394 hasConcept C104317684 @default.
• W2289096394 hasConcept C11413529 @default.
• W2289096394 hasConcept C121332964 @default.
• W2289096394 hasConcept C124681953 @default.
• W2289096394 hasConcept C134306372 @default.
• W2289096394 hasConcept C147168706 @default.
• W2289096394 hasConcept C154945302 @default.
• W2289096394 hasConcept C155281189 @default.
• W2289096394 hasConcept C162324750 @default.
• W2289096394 hasConcept C179254644 @default.
• W2289096394 hasConcept C185592680 @default.
• W2289096394 hasConcept C187736073 @default.
• W2289096394 hasConcept C18903297 @default.
• W2289096394 hasConcept C202444582 @default.
• W2289096394 hasConcept C204241405 @default.
• W2289096394 hasConcept C2777727622 @default.
• W2289096394 hasConcept C2778112365 @default.
• W2289096394 hasConcept C2780451532 @default.
• W2289096394 hasConcept C2986737658 @default.
• W2289096394 hasConcept C33923547 @default.
• W2289096394 hasConcept C35639132 @default.
• W2289096394 hasConcept C41008148 @default.
• W2289096394 hasConcept C50644808 @default.
• W2289096394 hasConcept C54355233 @default.
• W2289096394 hasConcept C55493867 @default.
• W2289096394 hasConcept C60644358 @default.
• W2289096394 hasConcept C74650414 @default.
• W2289096394 hasConcept C86803240 @default.
• W2289096394 hasConcept C90119067 @default.
• W2289096394 hasConceptScore W2289096394C104317684 @default.
• W2289096394 hasConceptScore W2289096394C11413529 @default.
• W2289096394 hasConceptScore W2289096394C121332964 @default.
• W2289096394 hasConceptScore W2289096394C124681953 @default.
• W2289096394 hasConceptScore W2289096394C134306372 @default.
• W2289096394 hasConceptScore W2289096394C147168706 @default.
• W2289096394 hasConceptScore W2289096394C154945302 @default.
• W2289096394 hasConceptScore W2289096394C155281189 @default.
• W2289096394 hasConceptScore W2289096394C162324750 @default.
• W2289096394 hasConceptScore W2289096394C179254644 @default.
• W2289096394 hasConceptScore W2289096394C185592680 @default.
• W2289096394 hasConceptScore W2289096394C187736073 @default.
• W2289096394 hasConceptScore W2289096394C18903297 @default.
• W2289096394 hasConceptScore W2289096394C202444582 @default.
• W2289096394 hasConceptScore W2289096394C204241405 @default.
• W2289096394 hasConceptScore W2289096394C2777727622 @default.
• W2289096394 hasConceptScore W2289096394C2778112365 @default.
• W2289096394 hasConceptScore W2289096394C2780451532 @default.
• W2289096394 hasConceptScore W2289096394C2986737658 @default.
• W2289096394 hasConceptScore W2289096394C33923547 @default.
• W2289096394 hasConceptScore W2289096394C35639132 @default.
• W2289096394 hasConceptScore W2289096394C41008148 @default.
• W2289096394 hasConceptScore W2289096394C50644808 @default.
• W2289096394 hasConceptScore W2289096394C54355233 @default.
• W2289096394 hasConceptScore W2289096394C55493867 @default.
• W2289096394 hasConceptScore W2289096394C60644358 @default.
• W2289096394 hasConceptScore W2289096394C74650414 @default.
• W2289096394 hasConceptScore W2289096394C86803240 @default.
• W2289096394 hasConceptScore W2289096394C90119067 @default.

• next