VLSI implementation of deep neural networks using integral stochastic computing

Arash Ardakani, Francois Leduc-Primeau, Naoya Onizawa, Takahiro Hanyu, Warren J. Gross

研究成果: Conference contribution

15 被引用数 (Scopus)

抄録

The hardware implementation of deep neural networks (DNNs) has recently received tremendous attention since many applications require high-speed operations. However, numerous processing elements and complex interconnections are usually required, leading to a large area occupation and a high power consumption. Stochastic computing has shown promising results for area-efficient hardware implementations, even though existing stochastic algorithms require long streams that exhibit long latency. In this paper, we propose an integer form of stochastic computation and introduce some elementary circuits. We then propose an efficient implementation of a DNN based on integral stochastic computing. The proposed architecture uses integer stochastic streams and a modified Finite State Machine-based tanh function to improve the performance and reduce the latency compared to existing stochastic architectures for DNN. The simulation results show the negligible performance loss of the proposed integer stochastic DNN for different network sizes compared to their floating point versions.

本文言語English
ホスト出版物のタイトル2016 9th International Symposium on Turbo Codes and Iterative Information Processing
ホスト出版物のサブタイトルPaths to 5G and Beyond, ISTC 2016
出版社IEEE Computer Society
ページ216-220
ページ数5
ISBN(電子版)9781509034017
DOI
出版ステータスPublished - 2016 10 17
イベント9th International Symposium on Turbo Codes and Iterative Information Processing, ISTC 2016 - Brest, France
継続期間: 2016 9 52016 9 9

出版物シリーズ

名前International Symposium on Turbo Codes and Iterative Information Processing, ISTC
2016-October
ISSN(印刷版)2165-4700
ISSN(電子版)2165-4719

Other

Other9th International Symposium on Turbo Codes and Iterative Information Processing, ISTC 2016
国/地域France
CityBrest
Period16/9/516/9/9

ASJC Scopus subject areas

  • 計算理論と計算数学
  • 情報システム
  • 理論的コンピュータサイエンス

フィンガープリント

「VLSI implementation of deep neural networks using integral stochastic computing」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル