يسعى هذا الموقع إلى تقديم معلومات علمية موثقة باللغة الإنجليزية عن كل ما يصدر باللغة العربية عن الحضارة الأوربية، وبذلك يسعى إلى تعريف الباحثين الغير ملمين بالعربية بالجهد العربي في هذا المجال. هذه المعلومات تأتى بالأساس من جمهورية مصر العربية، بدون إغفال أي إسهامات أخرى من الإخوة المتخصصين في الوطن العربي، شرقه وغربه. وكذلك يسعى الموقع إلى تقديم معلومات عن كل ما يصدر بلغات أوربية عن الحضارة العربية ولاسيما أثناء فترات النقل الثلاثة: نقلنا الأول عن اليونان (ما بين القرن الثاني والرابع الهجري/ الثامن والحادي عشر الميلادي)، ونقل الأوربيين عنا بلغتهم اللاتينية (ما بين القرن الخامس والتاسع الهجري/ الثاني عشر إلى الخامس عشر الميلادي) قبل نهضتهم الحديثة، والتي يردون بدايتها إلى القرن الخامس عشر الميلادي، ثم نقلنا الحديث عن الإنجليز والفرنسيين وغيرهم من الأوربيين بداية من الثاني عشر الهجري/القرن الثامن عشر الميلادي وحتى اليوم.

Saturday, February 25, 2017

Neural Machine Translation

[Retrieved from https://arxiv.org/abs/1609.08144 on 25.02.2017]

Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Yonghui Wu, Mike Schuster, Zhifeng Chen, Quoc V. Le, Mohammad Norouzi, Wolfgang Macherey, Maxim Krikun, Yuan Cao, Qin Gao, Klaus Macherey, Jeff Klingner, Apurva Shah, Melvin Johnson, Xiaobing Liu, Łukasz Kaiser, Stephan Gouws, Yoshikiyo Kato, Taku Kudo, Hideto Kazawa, Keith Stevens, George Kurian, Nishant Patil, Wei Wang, Cliff Young, Jason Smith, Jason Riesa, Alex Rudnick, Oriol Vinyals, Greg Corrado, Macduff Hughes, Jeffrey Dean
(Submitted on 26 Sep 2016 (v1), last revised 8 Oct 2016 (this version, v2))
Neural Machine Translation (NMT) is an end-to-end learning approach for automated translation, with the potential to overcome many of the weaknesses of conventional phrase-based translation systems. Unfortunately, NMT systems are known to be computationally expensive both in training and in translation inference. Also, most NMT systems have difficulty with rare words. These issues have hindered NMT's use in practical deployments and services, where both accuracy and speed are essential. In this work, we present GNMT, Google's Neural Machine Translation system, which attempts to address many of these issues. Our model consists of a deep LSTM network with 8 encoder and 8 decoder layers using attention and residual connections. To improve parallelism and therefore decrease training time, our attention mechanism connects the bottom layer of the decoder to the top layer of the encoder. To accelerate the final translation speed, we employ low-precision arithmetic during inference computations. To improve handling of rare words, we divide words into a limited set of common sub-word units ("wordpieces") for both input and output. This method provides a good balance between the flexibility of "character"-delimited models and the efficiency of "word"-delimited models, naturally handles translation of rare words, and ultimately improves the overall accuracy of the system. Our beam search technique employs a length-normalization procedure and uses a coverage penalty, which encourages generation of an output sentence that is most likely to cover all the words in the source sentence. On the WMT'14 English-to-French and English-to-German benchmarks, GNMT achieves competitive results to state-of-the-art. Using a human side-by-side evaluation on a set of isolated simple sentences, it reduces translation errors by an average of 60% compared to Google's phrase-based production system.

Subjects:Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Learning (cs.LG)
Cite as:arXiv:1609.08144 [cs.CL]
(or arXiv:1609.08144v2 [cs.CL] for this version)
Submission history
From: Mike Schuster [view email]
[v1] Mon, 26 Sep 2016 19:59:55 GMT (969kb,D)
[v2] Sat, 8 Oct 2016 19:10:41 GMT (968kb,D)
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
Link back to: arXiv, form interface, contact.

No comments:

Post a Comment