Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387029017> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4387029017 endingPage "14" @default.
- W4387029017 startingPage "1" @default.
- W4387029017 abstract "Generative Adversarial Networks (GANs) have been popularly researched in natural language generation, so-called Language GANs. Existing works adopt reinforcement learning (RL) based methods such as policy gradients for training Language GANs. The previous research of Language GANs usually focuses on stabilizing policy gradients or applying robust architectures (such as the large-scale pre-trained GPT-2) to achieve better performance. However, the quality and diversity of sampling are not guaranteed simultaneously. In this article, we propose a novel meta-learning-based generative adversarial network, Meta Exploration GAN (MetaEx-GAN), for ensuring the quality and diversity of sampling (sampling efficiency). In the proposed MetaEx-GAN, we develop an explorer trained by Meta Exploration to sample from the generated data to achieve better sampling efficiency. MetaEx-GAN employs MetaEx first applied to Language GANs to achieve better performance. We also propose a critical training method for MetaEx-GAN on the NLG task. According to our experimental results, MetaEx-GAN achieves state-of-the-art performance compared with existing Language GANs methods. Our experiments also demonstrate the generality of MetaEx-GAN with different architectures (involving GPT-2) and how MetaEx-GAN operates to improve Language GANs." @default.
- W4387029017 created "2023-09-26" @default.
- W4387029017 creator A5023752828 @default.
- W4387029017 creator A5040508737 @default.
- W4387029017 creator A5043595446 @default.
- W4387029017 creator A5057820524 @default.
- W4387029017 creator A5074764224 @default.
- W4387029017 date "2023-01-01" @default.
- W4387029017 modified "2023-09-26" @default.
- W4387029017 title "MetaEx-GAN: Meta Exploration to Improve Natural Language Generation via Generative Adversarial Networks" @default.
- W4387029017 doi "https://doi.org/10.1109/taslp.2023.3317571" @default.
- W4387029017 hasPublicationYear "2023" @default.
- W4387029017 type Work @default.
- W4387029017 citedByCount "0" @default.
- W4387029017 crossrefType "journal-article" @default.
- W4387029017 hasAuthorship W4387029017A5023752828 @default.
- W4387029017 hasAuthorship W4387029017A5040508737 @default.
- W4387029017 hasAuthorship W4387029017A5043595446 @default.
- W4387029017 hasAuthorship W4387029017A5057820524 @default.
- W4387029017 hasAuthorship W4387029017A5074764224 @default.
- W4387029017 hasConcept C108583219 @default.
- W4387029017 hasConcept C119857082 @default.
- W4387029017 hasConcept C127413603 @default.
- W4387029017 hasConcept C140779682 @default.
- W4387029017 hasConcept C154945302 @default.
- W4387029017 hasConcept C15744967 @default.
- W4387029017 hasConcept C195324797 @default.
- W4387029017 hasConcept C201995342 @default.
- W4387029017 hasConcept C2780451532 @default.
- W4387029017 hasConcept C2780767217 @default.
- W4387029017 hasConcept C2781002164 @default.
- W4387029017 hasConcept C2988773926 @default.
- W4387029017 hasConcept C37736160 @default.
- W4387029017 hasConcept C39890363 @default.
- W4387029017 hasConcept C41008148 @default.
- W4387029017 hasConcept C542102704 @default.
- W4387029017 hasConcept C76155785 @default.
- W4387029017 hasConcept C94915269 @default.
- W4387029017 hasConcept C97541855 @default.
- W4387029017 hasConceptScore W4387029017C108583219 @default.
- W4387029017 hasConceptScore W4387029017C119857082 @default.
- W4387029017 hasConceptScore W4387029017C127413603 @default.
- W4387029017 hasConceptScore W4387029017C140779682 @default.
- W4387029017 hasConceptScore W4387029017C154945302 @default.
- W4387029017 hasConceptScore W4387029017C15744967 @default.
- W4387029017 hasConceptScore W4387029017C195324797 @default.
- W4387029017 hasConceptScore W4387029017C201995342 @default.
- W4387029017 hasConceptScore W4387029017C2780451532 @default.
- W4387029017 hasConceptScore W4387029017C2780767217 @default.
- W4387029017 hasConceptScore W4387029017C2781002164 @default.
- W4387029017 hasConceptScore W4387029017C2988773926 @default.
- W4387029017 hasConceptScore W4387029017C37736160 @default.
- W4387029017 hasConceptScore W4387029017C39890363 @default.
- W4387029017 hasConceptScore W4387029017C41008148 @default.
- W4387029017 hasConceptScore W4387029017C542102704 @default.
- W4387029017 hasConceptScore W4387029017C76155785 @default.
- W4387029017 hasConceptScore W4387029017C94915269 @default.
- W4387029017 hasConceptScore W4387029017C97541855 @default.
- W4387029017 hasLocation W43870290171 @default.
- W4387029017 hasOpenAccess W4387029017 @default.
- W4387029017 hasPrimaryLocation W43870290171 @default.
- W4387029017 hasRelatedWork W2901368259 @default.
- W4387029017 hasRelatedWork W2919013397 @default.
- W4387029017 hasRelatedWork W2964157711 @default.
- W4387029017 hasRelatedWork W3156291593 @default.
- W4387029017 hasRelatedWork W3156763702 @default.
- W4387029017 hasRelatedWork W3159117918 @default.
- W4387029017 hasRelatedWork W3198184493 @default.
- W4387029017 hasRelatedWork W4206152009 @default.
- W4387029017 hasRelatedWork W4220812973 @default.
- W4387029017 hasRelatedWork W4313527397 @default.
- W4387029017 isParatext "false" @default.
- W4387029017 isRetracted "false" @default.
- W4387029017 workType "article" @default.