FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2947029787> ?p ?o ?g. }

• W2947029787 abstract "Can neural nets learn logic? We approach this classic question with current methods, and demonstrate that recurrent neural networks can learn to recognize first order logical entailment relations between expressions. We define an artificial language in first-order predicate logic, generate a large dataset of sample 'sentences', and use an automatic theorem prover to infer the relation between random pairs of such sentences. We describe a Siamese neural architecture trained to predict the logical relation, and experiment with recurrent and recursive networks. Siamese Recurrent Networks are surprisingly successful at the entailment recognition task, reaching near perfect performance on novel sentences (consisting of known words), and even outperforming recursive networks. We report a series of experiments to test the ability of the models to perform compositional generalization. In particular, we study how they deal with sentences of unseen length, and sentences containing unseen words. We show that set-ups using LSTMs and GRUs obtain high scores on these tests, demonstrating a form of compositionality." @default.
• W2947029787 created "2019-06-07" @default.
• W2947029787 creator A5007928903 @default.
• W2947029787 creator A5088379574 @default.
• W2947029787 date "2019-06-01" @default.
• W2947029787 modified "2023-09-23" @default.
• W2947029787 title "Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization." @default.
• W2947029787 cites W1699946128 @default.
• W2947029787 cites W174361228 @default.
• W2947029787 cites W1840435438 @default.
• W2947029787 cites W1889268436 @default.
• W2947029787 cites W1966678693 @default.
• W2947029787 cites W205765513 @default.
• W2947029787 cites W2064675550 @default.
• W2947029787 cites W2087451659 @default.
• W2947029787 cites W2110485445 @default.
• W2947029787 cites W2118373646 @default.
• W2947029787 cites W2118463056 @default.
• W2947029787 cites W2138913040 @default.
• W2947029787 cites W2144862731 @default.
• W2947029787 cites W2154770372 @default.
• W2947029787 cites W2157331557 @default.
• W2947029787 cites W215825141 @default.
• W2947029787 cites W2173051530 @default.
• W2947029787 cites W2250539671 @default.
• W2947029787 cites W2251869843 @default.
• W2947029787 cites W2293908967 @default.
• W2947029787 cites W2427182019 @default.
• W2947029787 cites W2508865106 @default.
• W2947029787 cites W2555308822 @default.
• W2947029787 cites W2781474777 @default.
• W2947029787 cites W2896556401 @default.
• W2947029787 cites W2900457592 @default.
• W2947029787 cites W2962765587 @default.
• W2947029787 cites W2962924847 @default.
• W2947029787 cites W2963267799 @default.
• W2947029787 cites W2963305465 @default.
• W2947029787 cites W2963440143 @default.
• W2947029787 cites W2963937837 @default.
• W2947029787 cites W3101885140 @default.
• W2947029787 cites W3137695714 @default.
• W2947029787 cites W6908809 @default.
• W2947029787 hasPublicationYear "2019" @default.
• W2947029787 type Work @default.
• W2947029787 sameAs 2947029787 @default.
• W2947029787 citedByCount "11" @default.
• W2947029787 countsByYear W29470297872019 @default.
• W2947029787 countsByYear W29470297872020 @default.
• W2947029787 countsByYear W29470297872021 @default.
• W2947029787 countsByYear W29470297872022 @default.
• W2947029787 crossrefType "posted-content" @default.
• W2947029787 hasAuthorship W2947029787A5007928903 @default.
• W2947029787 hasAuthorship W2947029787A5088379574 @default.
• W2947029787 hasConcept C100481476 @default.
• W2947029787 hasConcept C121375916 @default.
• W2947029787 hasConcept C134306372 @default.
• W2947029787 hasConcept C134752490 @default.
• W2947029787 hasConcept C140146324 @default.
• W2947029787 hasConcept C147168706 @default.
• W2947029787 hasConcept C154945302 @default.
• W2947029787 hasConcept C177148314 @default.
• W2947029787 hasConcept C199360897 @default.
• W2947029787 hasConcept C204321447 @default.
• W2947029787 hasConcept C25343380 @default.
• W2947029787 hasConcept C33923547 @default.
• W2947029787 hasConcept C41008148 @default.
• W2947029787 hasConcept C50644808 @default.
• W2947029787 hasConcept C77088390 @default.
• W2947029787 hasConcept C95318506 @default.
• W2947029787 hasConceptScore W2947029787C100481476 @default.
• W2947029787 hasConceptScore W2947029787C121375916 @default.
• W2947029787 hasConceptScore W2947029787C134306372 @default.
• W2947029787 hasConceptScore W2947029787C134752490 @default.
• W2947029787 hasConceptScore W2947029787C140146324 @default.
• W2947029787 hasConceptScore W2947029787C147168706 @default.
• W2947029787 hasConceptScore W2947029787C154945302 @default.
• W2947029787 hasConceptScore W2947029787C177148314 @default.
• W2947029787 hasConceptScore W2947029787C199360897 @default.
• W2947029787 hasConceptScore W2947029787C204321447 @default.
• W2947029787 hasConceptScore W2947029787C25343380 @default.
• W2947029787 hasConceptScore W2947029787C33923547 @default.
• W2947029787 hasConceptScore W2947029787C41008148 @default.
• W2947029787 hasConceptScore W2947029787C50644808 @default.
• W2947029787 hasConceptScore W2947029787C77088390 @default.
• W2947029787 hasConceptScore W2947029787C95318506 @default.
• W2947029787 hasOpenAccess W2947029787 @default.
• W2947029787 hasRelatedWork W2064675550 @default.
• W2947029787 hasRelatedWork W2118373646 @default.
• W2947029787 hasRelatedWork W2138708605 @default.
• W2947029787 hasRelatedWork W2210838531 @default.
• W2947029787 hasRelatedWork W2275485090 @default.
• W2947029787 hasRelatedWork W2411652523 @default.
• W2947029787 hasRelatedWork W2887970879 @default.
• W2947029787 hasRelatedWork W2889830342 @default.
• W2947029787 hasRelatedWork W2896154352 @default.
• W2947029787 hasRelatedWork W2949437134 @default.
• W2947029787 hasRelatedWork W2951278025 @default.
• W2947029787 hasRelatedWork W2962911926 @default.
• W2947029787 hasRelatedWork W2963005248 @default.
• W2947029787 hasRelatedWork W2963267799 @default.

• next