Matches in SemOpenAlex for { <https://semopenalex.org/work/W3176198369> ?p ?o ?g. }
Showing items 1 to 66 of
66
with 100 items per page.
- W3176198369 endingPage "12709" @default.
- W3176198369 startingPage "12701" @default.
- W3176198369 abstract "Natural language generation (NLG) is an important task with various applications like neural machine translation (NMT) and image captioning. Since deep-learning-based methods have issues of exposure bias and loss inconsistency, reinforcement learning (RL) is widely adopted in NLG tasks recently. But most RL-based methods ignore the deviation ignorance issue, which means the model fails to understand the extent of token-level deviation well. It leads to semantic incorrectness and hampers the agent to perform well. To address the issue, we propose a technique called adaptive prior-dependent correction (APDC) to enhance RL. It leverages the distribution generated by computing the distances between the ground truth and all other words to correct the agent's stochastic policy. Additionally, some techniques on RL are explored to coordinate RL with APDC, which requires a reward estimation at every time step. We find that the RL-based NLG tasks are a special case in RL, where the state transition is deterministic and the afterstate value equals the Q-value at every time step. To utilize such prior knowledge, we estimate the advantage function with the difference of the Q-values which can be estimated by Monte Carlo rollouts. Experiments show that, on three tasks of NLG (NMT, image captioning, abstractive text summarization), our method consistently outperforms the state-of-the-art RL-based approaches on different frequently-used metrics." @default.
- W3176198369 created "2021-07-05" @default.
- W3176198369 creator A5037724644 @default.
- W3176198369 creator A5081989147 @default.
- W3176198369 creator A5083652259 @default.
- W3176198369 date "2021-05-18" @default.
- W3176198369 modified "2023-09-27" @default.
- W3176198369 title "Adaptive Prior-Dependent Correction Enhanced Reinforcement Learning for Natural Language Generation" @default.
- W3176198369 doi "https://doi.org/10.1609/aaai.v35i14.17504" @default.
- W3176198369 hasPublicationYear "2021" @default.
- W3176198369 type Work @default.
- W3176198369 sameAs 3176198369 @default.
- W3176198369 citedByCount "1" @default.
- W3176198369 countsByYear W31761983692022 @default.
- W3176198369 crossrefType "journal-article" @default.
- W3176198369 hasAuthorship W3176198369A5037724644 @default.
- W3176198369 hasAuthorship W3176198369A5081989147 @default.
- W3176198369 hasAuthorship W3176198369A5083652259 @default.
- W3176198369 hasBestOaLocation W31761983691 @default.
- W3176198369 hasConcept C115961682 @default.
- W3176198369 hasConcept C119857082 @default.
- W3176198369 hasConcept C154945302 @default.
- W3176198369 hasConcept C157657479 @default.
- W3176198369 hasConcept C170858558 @default.
- W3176198369 hasConcept C195324797 @default.
- W3176198369 hasConcept C203005215 @default.
- W3176198369 hasConcept C204321447 @default.
- W3176198369 hasConcept C2776187449 @default.
- W3176198369 hasConcept C38652104 @default.
- W3176198369 hasConcept C41008148 @default.
- W3176198369 hasConcept C48145219 @default.
- W3176198369 hasConcept C97541855 @default.
- W3176198369 hasConceptScore W3176198369C115961682 @default.
- W3176198369 hasConceptScore W3176198369C119857082 @default.
- W3176198369 hasConceptScore W3176198369C154945302 @default.
- W3176198369 hasConceptScore W3176198369C157657479 @default.
- W3176198369 hasConceptScore W3176198369C170858558 @default.
- W3176198369 hasConceptScore W3176198369C195324797 @default.
- W3176198369 hasConceptScore W3176198369C203005215 @default.
- W3176198369 hasConceptScore W3176198369C204321447 @default.
- W3176198369 hasConceptScore W3176198369C2776187449 @default.
- W3176198369 hasConceptScore W3176198369C38652104 @default.
- W3176198369 hasConceptScore W3176198369C41008148 @default.
- W3176198369 hasConceptScore W3176198369C48145219 @default.
- W3176198369 hasConceptScore W3176198369C97541855 @default.
- W3176198369 hasIssue "14" @default.
- W3176198369 hasLocation W31761983691 @default.
- W3176198369 hasOpenAccess W3176198369 @default.
- W3176198369 hasPrimaryLocation W31761983691 @default.
- W3176198369 hasRelatedWork W2027280210 @default.
- W3176198369 hasRelatedWork W2122804826 @default.
- W3176198369 hasRelatedWork W2596543464 @default.
- W3176198369 hasRelatedWork W2747680751 @default.
- W3176198369 hasRelatedWork W3049752948 @default.
- W3176198369 hasRelatedWork W3093252849 @default.
- W3176198369 hasRelatedWork W3105439152 @default.
- W3176198369 hasRelatedWork W3176198369 @default.
- W3176198369 hasRelatedWork W4293797640 @default.
- W3176198369 hasRelatedWork W3045475294 @default.
- W3176198369 hasVolume "35" @default.
- W3176198369 isParatext "false" @default.
- W3176198369 isRetracted "false" @default.
- W3176198369 magId "3176198369" @default.
- W3176198369 workType "article" @default.