Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963559013> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2963559013 endingPage "798" @default.
- W2963559013 startingPage "788" @default.
- W2963559013 abstract "Sequence generative adversarial networks (SeqGAN) have been used to improve conditional sequence generation tasks, for example, chit-chat dialogue generation. To stabilize the training of SeqGAN, Monte Carlo tree search (MCTS) or reward at every generation step (REGS) is used to evaluate the goodness of a generated subsequence. MCTS is computationally intensive, but the performance of REGS is worse than MCTS. In this paper, we propose stepwise GAN (StepGAN), in which the discriminator is modified to automatically assign scores quantifying the goodness of each subsequence at every generation step. StepGAN has significantly less computational costs than MCTS. We demonstrate that StepGAN outperforms previous GAN-based methods on both synthetic experiment and chit-chat dialogue generation." @default.
- W2963559013 created "2019-07-30" @default.
- W2963559013 creator A5016343106 @default.
- W2963559013 creator A5040508737 @default.
- W2963559013 date "2019-04-01" @default.
- W2963559013 modified "2023-09-27" @default.
- W2963559013 title "Improving Conditional Sequence Generative Adversarial Networks by Stepwise Evaluation" @default.
- W2963559013 cites W2101105183 @default.
- W2963559013 cites W2119717200 @default.
- W2963559013 cites W2133012565 @default.
- W2963559013 cites W2136148671 @default.
- W2963559013 cites W2145339207 @default.
- W2963559013 cites W2581637843 @default.
- W2963559013 cites W2601324753 @default.
- W2963559013 cites W2604178507 @default.
- W2963559013 cites W2605045867 @default.
- W2963559013 cites W2963167310 @default.
- W2963559013 cites W2963206148 @default.
- W2963559013 cites W2963903950 @default.
- W2963559013 cites W635530177 @default.
- W2963559013 doi "https://doi.org/10.1109/taslp.2019.2896437" @default.
- W2963559013 hasPublicationYear "2019" @default.
- W2963559013 type Work @default.
- W2963559013 sameAs 2963559013 @default.
- W2963559013 citedByCount "39" @default.
- W2963559013 countsByYear W29635590132018 @default.
- W2963559013 countsByYear W29635590132019 @default.
- W2963559013 countsByYear W29635590132020 @default.
- W2963559013 countsByYear W29635590132021 @default.
- W2963559013 countsByYear W29635590132022 @default.
- W2963559013 countsByYear W29635590132023 @default.
- W2963559013 crossrefType "journal-article" @default.
- W2963559013 hasAuthorship W2963559013A5016343106 @default.
- W2963559013 hasAuthorship W2963559013A5040508737 @default.
- W2963559013 hasBestOaLocation W29635590132 @default.
- W2963559013 hasConcept C105795698 @default.
- W2963559013 hasConcept C113174947 @default.
- W2963559013 hasConcept C119857082 @default.
- W2963559013 hasConcept C134306372 @default.
- W2963559013 hasConcept C137877099 @default.
- W2963559013 hasConcept C154945302 @default.
- W2963559013 hasConcept C19499675 @default.
- W2963559013 hasConcept C2778112365 @default.
- W2963559013 hasConcept C2779803651 @default.
- W2963559013 hasConcept C33923547 @default.
- W2963559013 hasConcept C34388435 @default.
- W2963559013 hasConcept C37736160 @default.
- W2963559013 hasConcept C39890363 @default.
- W2963559013 hasConcept C41008148 @default.
- W2963559013 hasConcept C46149586 @default.
- W2963559013 hasConcept C54355233 @default.
- W2963559013 hasConcept C76155785 @default.
- W2963559013 hasConcept C86803240 @default.
- W2963559013 hasConcept C94915269 @default.
- W2963559013 hasConceptScore W2963559013C105795698 @default.
- W2963559013 hasConceptScore W2963559013C113174947 @default.
- W2963559013 hasConceptScore W2963559013C119857082 @default.
- W2963559013 hasConceptScore W2963559013C134306372 @default.
- W2963559013 hasConceptScore W2963559013C137877099 @default.
- W2963559013 hasConceptScore W2963559013C154945302 @default.
- W2963559013 hasConceptScore W2963559013C19499675 @default.
- W2963559013 hasConceptScore W2963559013C2778112365 @default.
- W2963559013 hasConceptScore W2963559013C2779803651 @default.
- W2963559013 hasConceptScore W2963559013C33923547 @default.
- W2963559013 hasConceptScore W2963559013C34388435 @default.
- W2963559013 hasConceptScore W2963559013C37736160 @default.
- W2963559013 hasConceptScore W2963559013C39890363 @default.
- W2963559013 hasConceptScore W2963559013C41008148 @default.
- W2963559013 hasConceptScore W2963559013C46149586 @default.
- W2963559013 hasConceptScore W2963559013C54355233 @default.
- W2963559013 hasConceptScore W2963559013C76155785 @default.
- W2963559013 hasConceptScore W2963559013C86803240 @default.
- W2963559013 hasConceptScore W2963559013C94915269 @default.
- W2963559013 hasFunder F4320322795 @default.
- W2963559013 hasIssue "4" @default.
- W2963559013 hasLocation W29635590131 @default.
- W2963559013 hasLocation W29635590132 @default.
- W2963559013 hasOpenAccess W2963559013 @default.
- W2963559013 hasPrimaryLocation W29635590131 @default.
- W2963559013 hasRelatedWork W2412510955 @default.
- W2963559013 hasRelatedWork W2564929713 @default.
- W2963559013 hasRelatedWork W2786039921 @default.
- W2963559013 hasRelatedWork W2809882560 @default.
- W2963559013 hasRelatedWork W2952541330 @default.
- W2963559013 hasRelatedWork W2952936466 @default.
- W2963559013 hasRelatedWork W3005996785 @default.
- W2963559013 hasRelatedWork W3119931323 @default.
- W2963559013 hasRelatedWork W4280544492 @default.
- W2963559013 hasRelatedWork W4293320219 @default.
- W2963559013 hasVolume "27" @default.
- W2963559013 isParatext "false" @default.
- W2963559013 isRetracted "false" @default.
- W2963559013 magId "2963559013" @default.
- W2963559013 workType "article" @default.