18233

Marian: Cost-effective High-Quality Neural Machine Translation in C++

Marcin Junczys-Dowmunt, Kenneth Heafield, Hieu Hoang, Roman Grundkiewicz, Anthony Aue
Microsoft Translator, 1 Microsoft Way, Redmond, WA 98121, USA
arXiv:1805.12096 [cs.CL], (30 May 2018)

@article{junczys-dowmunt2018marian,

   title={Marian: Cost-effective High-Quality Neural Machine Translation in C++},

   author={Junczys-Dowmunt, Marcin and Heafield, Kenneth and Hoang, Hieu and Grundkiewicz, Roman and Aue, Anthony},

   year={2018},

   month={may},

   archivePrefix={"arXiv"},

   primaryClass={cs.CL}

}

This paper describes the submissions of the "Marian" team to the WNMT 2018 shared task. We investigate combinations of teacher-student training, low-precision matrix products, auto-tuning and other methods to optimize the Transformer model on GPU and CPU. By further integrating these methods with the new averaging attention networks, a recently introduced faster Transformer variant, we create a number of high-quality, high-performance models on the GPU and CPU, dominating the Pareto frontier for this shared task.
Rating: 3.7/5. From 3 votes.
Please wait...

* * *

* * *

HGPU group © 2010-2018 hgpu.org

All rights belong to the respective authors

Contact us: