Efficient methods for mapping neural machine translator on FPGAs
Li, Qin
Loading…
Permalink
https://hdl.handle.net/2142/108193
Description
Title
Efficient methods for mapping neural machine translator on FPGAs
Author(s)
Li, Qin
Issue Date
2020-05-13
Director of Research (if dissertation) or Advisor (if thesis)
Chen, Deming
Department of Study
Electrical & Computer Eng
Discipline
Electrical & Computer Engr
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
M.S.
Degree Level
Thesis
Keyword(s)
FPGA, HLS
Abstract
Neural machine translation (NMT) is one of the most critical applications in natural language processing (NLP) with the main idea to convert text in one language to another language using deep neural networks. In recent year, we have seen continuous development of NMT by integrating more emerging technologies, such as bidirectional gated recurrent units (GRU), attention mechanisms, and beam-search algorithms, for improved translation quality. However, with the increasing problem size, the real-life NMT models have become much more complicated and difficult to implement on hardware for acceleration opportunities. In this thesis, we aim to exploit the capability of FPGAs for delivering highly efficient implementations for real-life NMT applications. In our work, we map the inference of a large-scale NMT model with total computation of 172 GFLOP to a highly optimized high-level synthesis (HLS) IP and integrate the IP into Xilinx VCU118 FPGA platform. The model has widely used key features for NMTs including bidirectional GRU layer, attention mechanism, and beam search algorithm. We quantize the model to mixed-precision representation in which parameters and portions of calculations are in 16-bit half precision, and others remain as 32-bit floating-point. Compared to the float NMT implementation on FPGA, we achieve 13.1x speedup with end-to-end performance of 22.0 GFLOPS without any accuracy degradation.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
