Abstract | ||
---|---|---|
Back-translation has become a commonly employed heuristic for semi-supervised neural machine translation. The technique is both straightforward to apply and has led to state-of-the-art results. In this work, we offer a principled interpretation of back-translation as approximate inference in a generative model of bitext and show how the standard implementation of back-translation corresponds to a single iteration of the wake-sleep algorithm in our proposed model. Moreover, this interpretation suggests a natural iterative generalization, which we demonstrate leads to further improvement of up to 1.6 BLEU. |
Year | Venue | Field |
---|---|---|
2018 | arXiv: Computation and Language | Wake,BLEU,Heuristic,Back translation,Computer science,Generalization,Machine translation,Approximate inference,Artificial intelligence,Machine learning,Generative model |
DocType | Volume | Citations |
Journal | abs/1806.04402 | 2 |
PageRank | References | Authors |
0.37 | 17 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ryan Cotterell | 1 | 85 | 13.66 |
julia kreutzer | 2 | 22 | 5.92 |