TextSearch

Abstract page for arXiv paper 1409.0473: Neural Machine Translation by Jointly Learning to Align and Translate

View PDF Abstract:Neural machine translation is a recently proposed approach to machine translation. Unlike the traditional statistical machine translation, the neural machine translation aims at building a single neural network that can be jointly tuned to maximize the translation performance. The models proposed recently for neural machine translation often belong to a family of encoder-decoders and consists of an encoder that encodes a source sentence into a fixed-length vector from which a decoder generates a translation. In this paper, we conjecture that the use of a fixed-length vector is a bottleneck in improving the performance of this basic encoder-decoder architecture, and propose to extend this by allowing a model to automatically (soft-)search for parts of a source sentence that are relevant to predicting a target word, without having to form these parts as a hard segment explicitly. With this new approach, we achieve a translation performance comparable to the existing state-of-the-art phrase-based system on the task of English-to-French translation. Furthermore, qualitative analysis reveals that the (soft-)alignments found by the model agree well with our intuition. Submission history From: Dzmitry Bahdanau [view email] [v1] Mon, 1 Sep 2014 16:33:02 UTC (83 KB) [v2] Thu, 4 Sep 2014 18:32:00 UTC (83 KB) [v3] Tue, 7 Oct 2014 18:10:39 UTC (84 KB) [v4] Fri, 19 Dec 2014 21:39:11 UTC (107 KB) [v5] Sun, 22 Mar 2015 17:08:39 UTC (107 KB) [v6] Fri, 24 Apr 2015 13:25:33 UTC (144 KB) [v7] Thu, 19 May 2016 21:53:22 UTC (144 KB)

[1409.0473] Neural Machine Translation by Jointly Learning to Align and Translate