• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Optimum Data Representation and Its Accuracy Assurance for Reconfigurable Accelerators

Research Project

Project/Area Number 18H03217
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 60040:Computer system-related
Research InstitutionWaseda University

Principal Investigator

Kimura Shinji  早稲田大学, 理工学術院(情報生産システム研究科・センター), 教授 (20183303)

Co-Investigator(Kenkyū-buntansha) 戸川 望  早稲田大学, 理工学術院, 教授 (30298161)
Project Period (FY) 2018-04-01 – 2021-03-31
Project Status Completed (Fiscal Year 2020)
Budget Amount *help
¥17,550,000 (Direct Cost: ¥13,500,000、Indirect Cost: ¥4,050,000)
Fiscal Year 2020: ¥3,380,000 (Direct Cost: ¥2,600,000、Indirect Cost: ¥780,000)
Fiscal Year 2019: ¥4,940,000 (Direct Cost: ¥3,800,000、Indirect Cost: ¥1,140,000)
Fiscal Year 2018: ¥9,230,000 (Direct Cost: ¥7,100,000、Indirect Cost: ¥2,130,000)
Keywordsエラー耐性に基づく最適化 / 誤差解析 / Approximate Computing / 誤差と計算結果の関係 / 近似計算 / データ表現形式 / 共有指数表現 / データ表現と誤差解析 / データ圧縮 / 再構成可能アーキテクチャ / データ表現と精度保証
Outline of Final Research Achievements

The project is on the optimum data representation and its accuracy assurance for reconfigurable accelerators including reconfigurable hardware modules such as FPGA (Field Programmable Logic Array). A reconfigurable accelerator can construct dedicated special hardware accelerators depending on applications. In the optimization of data representation for reconfigurable accelerators, the area, delay and power are optimized under the error tolerance of applications. On image processing and image recognition applications, new data representation methods, operational units for the data representation, and their evaluation methods have been devised and evaluated.

Academic Significance and Societal Importance of the Research Achievements

近年、CNN (Convolutional Neural Network, 畳み込みニューラルネットワーク)のように、非常に多くの演算を必要とする応用が用いられるようになってきた。そのハードウェアによる高速化は実応用においては非常に重要であり、端末側からサーバー側まで広くハードウェアアクセラレータが用いられている。再構成アクセラレータはそのような応用志向のハードウェアを実現するプラットフォームであり、本プロジェクトで、実際に再構成アクセラレータへ向けたデータ表現とその誤差評価の手法や演算器の提案を行ったことは、学術上および実用上の意義が高い。

Report

(4 results)
  • 2020 Annual Research Report   Final Research Report ( PDF )
  • 2019 Annual Research Report
  • 2018 Annual Research Report
  • Research Products

    (11 results)

All 2020 2019 2018

All Journal Article (3 results) (of which Int'l Joint Research: 3 results,  Peer Reviewed: 3 results,  Open Access: 1 results) Presentation (8 results) (of which Int'l Joint Research: 7 results)

  • [Journal Article] Approximate FPGA-Based Multipliers Using Carry-Inexact Elementary Modules2020

    • Author(s)
      Yi GUO, Heming SUN, Ping LEI, Shinji KIMURA
    • Journal Title

      IEICE Transactions on Fundamentals

      Volume: E103A Issue: 9 Pages: 1054-1062

    • DOI

      10.1587/transfun.2019kep0002

    • NAID

      130007893715

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Design of Low-Cost Approximate Multipliers Based on Probability-Driven Inexact Compressors2019

    • Author(s)
      GUO Yi、SUN Heming、LEI Ping、KIMURA Shinji
    • Journal Title

      IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

      Volume: E102.A Issue: 12 Pages: 1781-1791

    • DOI

      10.1587/transfun.E102.A.1781

    • NAID

      130007754045

    • ISSN
      0916-8508, 1745-1337
    • Year and Date
      2019-12-01
    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Lossy Compression for Embedded Computer Vision Systems2018

    • Author(s)
      Li Guo, Dajiang Zhou, Jinjia Zhou, Shinji Kimura, and Satoshi Goto
    • Journal Title

      IEEE Access

      Volume: 6 Pages: 39385-39397

    • DOI

      10.1109/access.2018.2852809

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Presentation] Accuracy-Configurable Low-Power Approximate Floating-Point Multiplier Based on Mantissa Bit Segmentation2020

    • Author(s)
      Jie Li, Yi Guo and Shinji Kimura
    • Organizer
      IEEE Region 10 Conference (TENCON)
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Approximate Floating Point Multiplier based on Shifting Addition Using Carry Signal from Second-Highest-Bit2020

    • Author(s)
      Jie LI, Yi GUO, and Shinji KIMURA
    • Organizer
      IEICE Tech. Report, VLD2019-120
    • Related Report
      2019 Annual Research Report
  • [Presentation] Small-Area and Low-Power FPGA-Based Multipliers using Approximate Elementary Modules2020

    • Author(s)
      Guo Yi、Sun Heming、Kimura Shinji
    • Organizer
      Proc. of ASP-DAC 2020
    • Related Report
      2019 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Embedded Frame Compression for Energy-Efficient Computer Vision Systems2018

    • Author(s)
      Li Guo, Dajiang Zhou, Jinjia Zhou, Shinji Kimura
    • Organizer
      ISCAS 2018
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Sparseness Ratio Allocation and Neuron Re-pruning for Neural Networks Compression2018

    • Author(s)
      Li Guo, Dajiang Zhou, Jinjia Zhou, Shinji Kimura
    • Organizer
      ISCAS 2018
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Low Cost Approximate Multiplier Design using Probability Driven Inexac t Compressors2018

    • Author(s)
      Yi Guo, Heming Sun, Li Guo, Shinji Kimura
    • Organizer
      APCCAS 2018
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Energy-Efficient and High Performance Approximate Multiplier Using Compre ssors Based on Input Reordering2018

    • Author(s)
      Zhenhao Liu, Yi Guo, Xiaoting Sun and Shinji Kimura
    • Organizer
      TENCON 2018
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] A Radix-4 Partial Product Generation-Based Approximate Multiplier for High-Speed and Low-Power Digital Signal Processing2018

    • Author(s)
      Xiaoting Sun, Yi Guo, Zhenhao Liu, Shinji Kimura
    • Organizer
      ICECS 2018
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research

URL: 

Published: 2018-04-23   Modified: 2022-01-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi