File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Appears in Collections:
Conference Paper: Neural machine translation with Gumbel-Greedy Decoding
Title | Neural machine translation with Gumbel-Greedy Decoding |
---|---|
Authors | |
Keywords | Machine Translation Gumbel Softmax Greedy Decoding Generator Discriminator |
Issue Date | 2018 |
Publisher | Association for the Advancement of Artificial Intelligence (AAAI) Press. |
Citation | Proceedings of the Thirty-Second Association for the Advancement of Artificial Intelligence (AAAI) Conference on Artificial Intelligence (AAAI-18), New Orleans, Louisiana, USA, 2-7 February 2018, p. 5125-5132 How to Cite? |
Abstract | Previous neural machine translation models used some heuristic search algorithms (e.g., beam search) in order to avoid solving the maximum a posteriori problem over translation sentences at test phase. In this paper, we propose the extit{Gumbel-Greedy Decoding} which trains a generative network to predict translation under a trained model. We solve such a problem using the Gumbel-Softmax reparameterization, which makes our generative network differentiable and trainable through standard stochastic gradient methods. We empirically demonstrate that our proposed model is effective for generating sequences of discrete words. |
Description | Session: AAAI18 - NLP and Machine Learning |
Persistent Identifier | http://hdl.handle.net/10722/262422 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Gu, J | - |
dc.contributor.author | Im, DJ | - |
dc.contributor.author | Li, VOK | - |
dc.date.accessioned | 2018-09-28T04:59:04Z | - |
dc.date.available | 2018-09-28T04:59:04Z | - |
dc.date.issued | 2018 | - |
dc.identifier.citation | Proceedings of the Thirty-Second Association for the Advancement of Artificial Intelligence (AAAI) Conference on Artificial Intelligence (AAAI-18), New Orleans, Louisiana, USA, 2-7 February 2018, p. 5125-5132 | - |
dc.identifier.uri | http://hdl.handle.net/10722/262422 | - |
dc.description | Session: AAAI18 - NLP and Machine Learning | - |
dc.description.abstract | Previous neural machine translation models used some heuristic search algorithms (e.g., beam search) in order to avoid solving the maximum a posteriori problem over translation sentences at test phase. In this paper, we propose the extit{Gumbel-Greedy Decoding} which trains a generative network to predict translation under a trained model. We solve such a problem using the Gumbel-Softmax reparameterization, which makes our generative network differentiable and trainable through standard stochastic gradient methods. We empirically demonstrate that our proposed model is effective for generating sequences of discrete words. | - |
dc.language | eng | - |
dc.publisher | Association for the Advancement of Artificial Intelligence (AAAI) Press. | - |
dc.relation.ispartof | AAAI Conference on Artificial Intelligence, AAAI-18 | - |
dc.subject | Machine Translation | - |
dc.subject | Gumbel Softmax | - |
dc.subject | Greedy Decoding | - |
dc.subject | Generator Discriminator | - |
dc.title | Neural machine translation with Gumbel-Greedy Decoding | - |
dc.type | Conference_Paper | - |
dc.identifier.email | Li, VOK: vli@eee.hku.hk | - |
dc.identifier.authority | Li, VOK=rp00150 | - |
dc.identifier.hkuros | 292177 | - |
dc.identifier.spage | 5125 | - |
dc.identifier.epage | 5132 | - |
dc.publisher.place | United States | - |