Name
Abstract | ||
|---|---|---|
•The traditional neural machine translation is typically optimized with human-designed approximate loss function.•The proposed method trains the neural machine translation model to generate human-like translations directly.•The single generator-discriminator net is extended into the multi adversarial net which contains multiple generators and discriminators.•The proposed models achieve significant improvements. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1016/j.neucom.2018.09.006 | Neurocomputing |
Keywords | Field | DocType |
Neural machine translation,Multi generative adversarial net,Human-like translation | Minimax,Discriminator,Machine translation,Artificial intelligence,Generative grammar,Nash equilibrium,Discriminative model,Sentence,Mathematics,Machine learning,Generative model | Journal |
Volume | ISSN | Citations |
321 | 0925-2312 | 1 |
PageRank | References | Authors |
0.35 | 22 | 4 |