Share Email Print

Proceedings Paper

Improving neural network performance on SIMD architectures
Author(s): Elena Limonova; Dmitry Ilin; Dmitry Nikolaev
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Neural network calculations for the image recognition problems can be very time consuming. In this paper we propose three methods of increasing neural network performance on SIMD architectures. The usage of SIMD extensions is a way to speed up neural network processing available for a number of modern CPUs. In our experiments, we use ARM NEON as SIMD architecture example. The first method deals with half float data type for matrix computations. The second method describes fixed-point data type for the same purpose. The third method considers vectorized activation functions implementation. For each method we set up a series of experiments for convolutional and fully connected networks designed for image recognition task.

Paper Details

Date Published: 8 December 2015
PDF: 6 pages
Proc. SPIE 9875, Eighth International Conference on Machine Vision (ICMV 2015), 98750L (8 December 2015); doi: 10.1117/12.2228594
Show Author Affiliations
Elena Limonova, Moscow Institute of Physics and Technology (Russian Federation)
Dmitry Ilin, Smart Engines Ltd. (Russian Federation)
Dmitry Nikolaev, Institute For Information Transmission Problems (Russian Federation)

Published in SPIE Proceedings Vol. 9875:
Eighth International Conference on Machine Vision (ICMV 2015)
Antanas Verikas; Petia Radeva; Dmitry Nikolaev, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?