Matches in SemOpenAlex for { <https://semopenalex.org/work/W3185383710> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W3185383710 abstract "We present the Power Law Graph Transformer, a transformer model with well defined deductive and inductive tasks for prediction and representation learning. The deductive task learns the dataset level (global) and instance level (local) graph structures in terms of learnable power law distribution parameters. The inductive task outputs the prediction probabilities using the deductive task output, similar to a transductive model. We trained our model with Turkish-English and Portuguese-English datasets from TED talk transcripts for machine translation and compared the model performance and characteristics to a transformer model with scaled dot product attention trained on the same experimental setup. We report BLEU scores of $17.79$ and $28.33$ on the Turkish-English and Portuguese-English translation tasks with our model, respectively. We also show how a duality between a quantization set and N-dimensional manifold representation can be leveraged to transform between local and global deductive-inductive outputs using successive application of linear and non-linear transformations end-to-end." @default.
- W3185383710 created "2021-08-02" @default.
- W3185383710 creator A5025289341 @default.
- W3185383710 date "2021-06-27" @default.
- W3185383710 modified "2023-09-27" @default.
- W3185383710 title "Power Law Graph Transformer for Machine Translation and Representation Learning." @default.
- W3185383710 cites W1522301498 @default.
- W3185383710 cites W1533861849 @default.
- W3185383710 cites W1902237438 @default.
- W3185383710 cites W2095705004 @default.
- W3185383710 cites W2101105183 @default.
- W3185383710 cites W2121879602 @default.
- W3185383710 cites W2132339004 @default.
- W3185383710 cites W2133564696 @default.
- W3185383710 cites W2153579005 @default.
- W3185383710 cites W2250539671 @default.
- W3185383710 cites W2525778437 @default.
- W3185383710 cites W2594990650 @default.
- W3185383710 cites W2626778328 @default.
- W3185383710 cites W2912225506 @default.
- W3185383710 cites W2950577311 @default.
- W3185383710 cites W2950635152 @default.
- W3185383710 cites W2963341956 @default.
- W3185383710 cites W2963532001 @default.
- W3185383710 cites W2964085268 @default.
- W3185383710 cites W2994617532 @default.
- W3185383710 cites W3037932933 @default.
- W3185383710 hasPublicationYear "2021" @default.
- W3185383710 type Work @default.
- W3185383710 sameAs 3185383710 @default.
- W3185383710 citedByCount "0" @default.
- W3185383710 crossrefType "posted-content" @default.
- W3185383710 hasAuthorship W3185383710A5025289341 @default.
- W3185383710 hasConcept C119857082 @default.
- W3185383710 hasConcept C121332964 @default.
- W3185383710 hasConcept C132525143 @default.
- W3185383710 hasConcept C154945302 @default.
- W3185383710 hasConcept C165801399 @default.
- W3185383710 hasConcept C203005215 @default.
- W3185383710 hasConcept C204321447 @default.
- W3185383710 hasConcept C41008148 @default.
- W3185383710 hasConcept C62520636 @default.
- W3185383710 hasConcept C66322947 @default.
- W3185383710 hasConcept C80444323 @default.
- W3185383710 hasConceptScore W3185383710C119857082 @default.
- W3185383710 hasConceptScore W3185383710C121332964 @default.
- W3185383710 hasConceptScore W3185383710C132525143 @default.
- W3185383710 hasConceptScore W3185383710C154945302 @default.
- W3185383710 hasConceptScore W3185383710C165801399 @default.
- W3185383710 hasConceptScore W3185383710C203005215 @default.
- W3185383710 hasConceptScore W3185383710C204321447 @default.
- W3185383710 hasConceptScore W3185383710C41008148 @default.
- W3185383710 hasConceptScore W3185383710C62520636 @default.
- W3185383710 hasConceptScore W3185383710C66322947 @default.
- W3185383710 hasConceptScore W3185383710C80444323 @default.
- W3185383710 hasLocation W31853837101 @default.
- W3185383710 hasOpenAccess W3185383710 @default.
- W3185383710 hasPrimaryLocation W31853837101 @default.
- W3185383710 hasRelatedWork W1503463901 @default.
- W3185383710 hasRelatedWork W1503527899 @default.
- W3185383710 hasRelatedWork W1537765963 @default.
- W3185383710 hasRelatedWork W1552228685 @default.
- W3185383710 hasRelatedWork W1561384305 @default.
- W3185383710 hasRelatedWork W1564248646 @default.
- W3185383710 hasRelatedWork W1582731059 @default.
- W3185383710 hasRelatedWork W1841605138 @default.
- W3185383710 hasRelatedWork W1894395204 @default.
- W3185383710 hasRelatedWork W2089745520 @default.
- W3185383710 hasRelatedWork W2145474226 @default.
- W3185383710 hasRelatedWork W2725430220 @default.
- W3185383710 hasRelatedWork W2806373217 @default.
- W3185383710 hasRelatedWork W2903639586 @default.
- W3185383710 hasRelatedWork W2997932512 @default.
- W3185383710 hasRelatedWork W2998702685 @default.
- W3185383710 hasRelatedWork W3007859710 @default.
- W3185383710 hasRelatedWork W3036428880 @default.
- W3185383710 hasRelatedWork W3092916550 @default.
- W3185383710 hasRelatedWork W3199755688 @default.
- W3185383710 isParatext "false" @default.
- W3185383710 isRetracted "false" @default.
- W3185383710 magId "3185383710" @default.
- W3185383710 workType "article" @default.