6533b86dfe1ef96bd12c973a

RESEARCH PRODUCT

Fast Neural Machine Translation Implementation

Tomasz DwojakDaniel TorregrosaKenneth HeafieldRihards KrišlauksHieu Hoang

subject

FOS: Computer and information sciencesFocus (computing)Computer Science - Computation and LanguageMachine translationComputer sciencebusiness.industrycomputer.software_genreTrack (rail transport)Softmax functionArtificial intelligenceInference enginebusinesscomputerComputation and Language (cs.CL)

description

This paper describes the submissions to the efficiency track for GPUs at the Workshop for Neural Machine Translation and Generation by members of the University of Edinburgh, Adam Mickiewicz University, Tilde and University of Alicante. We focus on efficient implementation of the recurrent deep-learning model as implemented in Amun, the fast inference engine for neural machine translation. We improve the performance with an efficient mini-batching algorithm, and by fusing the softmax operation with the k-best extraction algorithm. Submissions using Amun were first, second and third fastest in the GPU efficiency track.

yearjournalcountryeditionlanguage
2018-01-01
http://arxiv.org/abs/1805.09863