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

Development of a Probabilistic Sampling Machine for Spectral Decomposition using FPGA

Research Project

Project/Area Number 19K12154
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 61040:Soft computing-related
Research InstitutionThe University of Electro-Communications

Principal Investigator

Shouno Hayaru  電気通信大学, 大学院情報理工学研究科, 教授 (50263231)

Project Period (FY) 2019-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2021: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2020: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2019: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Keywordsエッジデバイス / FPGA / MCMC法 / スペクトル分解 / MCMC / X線分子分光法 / モンテカルロ法 / 温度交換モンテカルロ法 / 並列計算
Outline of Research at the Start

統計的機械学習におけるベイズ推論において,サンプリングによる期待値の推定精度は学習機械の性能を大きく左右する因子である.一方,大規模な機械学習モデルでは,サンプリングの計算量が問題になる.本研究課題では,FPGAデバイスを用いた数値計算システムを構築し,その上で交換モンテカルロ法を動かす並列システムを構築する.この並列システムを現実に取得される,スペクトル分解を中心とした実問題を例題に,簡便なハードウェアによるアクセラレーターシステムを構築することを目指す.

Outline of Final Research Achievements

This project aims to verify the feasibility of implementing computationally expensive models on FPGAs. The main objectives of this project include evaluating computational costs and programming the Markov Chain Monte Carlo (MCMC) method for FPGA implementation. Specifically, we implemented a computational model based on the Ising spin model on an FPGA and applied the MCMC method to compute the states of this model. Additionally, to apply this method to spectral decomposition, we developed a computational algorithm that combines the temperature-exchange MCMC method and evolutionary computation for spectral decomposition.

Academic Significance and Societal Importance of the Research Achievements

省電力で計算コストが安価なFPGAはエッジデバイスとして,今後の計算機発展において重要な役割を果たすと考えられる.本研究ではFPGAを中心として,実問題へアプローチするために計算モデルを簡略化してどの程度の成果が得られるのかを試行している.その結果,計算モデルにおけるビット演算精度を落とすといった工夫を用いることにより,一定の計算精度を担保した形で,深層学習やマルコフチェーンモンテカルロ法などの計算手法がFPGA上で実現できることを示した.

Report

(5 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Research-status Report
  • 2020 Research-status Report
  • 2019 Research-status Report
  • Research Products

    (17 results)

All 2022 2021 2020 2019

All Journal Article (8 results) (of which Peer Reviewed: 8 results,  Open Access: 5 results) Presentation (9 results) (of which Int'l Joint Research: 5 results)

  • [Journal Article] Improving sign-algorithm convergence rate using natural gradient for lossless audio compression2022

    • Author(s)
      Mineo Taiyo、Shouno Hayaru
    • Journal Title

      EURASIP Journal on Audio, Speech, and Music Processing

      Volume: 2022 Issue: 1 Pages: 1-15

    • DOI

      10.1186/s13636-022-00243-w

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Prediction of metal temperature by microstructural features in creep exposed austenitic stainless steel with sparse modeling2021

    • Author(s)
      Endo Akihiro、Sawada Kota、Nagata Kenji、Yoshikawa Hideki、Shouno Hayaru
    • Journal Title

      Science and Technology of Advanced Materials: Methods

      Volume: 1 Issue: 1 Pages: 225-233

    • DOI

      10.1080/27660400.2021.1997556

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Determination of common peak structure from multiple X-ray photo-electron spectroscopy data sets2021

    • Author(s)
      Murakami Ryo、Shouno Hayaru、Nagata Kenji、Shinotsuka Hiroshi、Yoshikawa Hideki
    • Journal Title

      Science and Technology of Advanced Materials: Methods

      Volume: 1 Issue: 1 Pages: 182-191

    • DOI

      10.1080/27660400.2021.1957304

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Bayesian estimation for XPS spectral analysis at multiple core levels2021

    • Author(s)
      Machida Atsushi、Nagata Kenji、Murakami Ryo、Shinotsuka Hiroshi、Shouno Hayaru、Yoshikawa Hideki、Okada Masato
    • Journal Title

      Science and Technology of Advanced Materials: Methods

      Volume: 1 Issue: 1 Pages: 123-133

    • DOI

      10.1080/27660400.2021.1943172

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Development of multiple core-level XPS spectra decomposition method based on the Bayesian information criterion2020

    • Author(s)
      Murakami Ryo、Tanaka Hiromi、Shinotsuka Hiroshi、Nagata Kenji、Shouno Hayaru、Yoshikawa Hideki
    • Journal Title

      Journal of Electron Spectroscopy and Related Phenomena

      Volume: 245 Pages: 147003-147003

    • DOI

      10.1016/j.elspec.2020.147003

    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Journal Article] Development of spectral decomposition based on Bayesian information criterion with estimation of confidence interval2020

    • Author(s)
      Shinotsuka Hiroshi、Nagata Kenji、Yoshikawa Hideki、Mototake Yoh-Ichi、Shouno Hayaru、Okada Masato
    • Journal Title

      Science and Technology of Advanced Materials

      Volume: 21 Issue: 1 Pages: 402-419

    • DOI

      10.1080/14686996.2020.1773210

    • Related Report
      2020 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] ベイズ的変数選択に基づく分光スペクトル分解2019

    • Author(s)
      川島 貴大 , 庄野 逸
    • Journal Title

      情報処理学会論文誌数理モデル化と応用

      Volume: 12 Pages: 34-43

    • NAID

      170000150459

    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Journal Article] Bolassoを用いたびまん性肺疾患画像の特徴選択2019

    • Author(s)
      遠藤 瑛泰 , 永田 賢二 , 木戸 尚治 , 庄野 逸
    • Journal Title

      情報処理学会論文誌数理モデル化と応用

      Volume: 12 Pages: 68-77

    • NAID

      170000181266

    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Presentation] Magnetic Domain Texture Pattern Analysis Using Wavelet-Based Joint Texture Statistics2021

    • Author(s)
      Ryo Murakami, Masaichiro Mizumaki, Ichiro Akai, Hayaru Shouno
    • Organizer
      Materials Research Meeting 21
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Prediction of metal temperature in creep-exposed austenitic stainless steel from optical micrographs with sparse regression method2021

    • Author(s)
      Akihiro Endo, Kota Sawada, Kenji Nagata, Hideki Yoshikawa, Hayaru Shouno,
    • Organizer
      Materials Research Meeting 21
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Bayesian Dynamic Mode Decomposition with Variational Matrix Factorization2021

    • Author(s)
      Takahiro Kawashima, Hayaru Shouno, Hideitsu Hino
    • Organizer
      Conference on Artificial Intelligence (AAAI-21)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Bayesian Sparse Covariance Structure Analysis for Correlated Count Data2020

    • Author(s)
      Sho Ichigozaki, Takahiro Kawashima, Hayaru Shouno
    • Organizer
      Parallel and Distributed Processing Techniques and Applications
    • Related Report
      2020 Research-status Report
  • [Presentation] 参照スペクトルを使った多元素XPSスペクトルの解析手法の開発2020

    • Author(s)
      村上 諒, 庄野 逸, 永田 賢二, 篠塚 寛志, 吉川 英樹
    • Organizer
      2020年日本表面真空学会学術講演会
    • Related Report
      2020 Research-status Report
  • [Presentation] 多量のスペクトルデータを利用した参照スペクトルの推定手法の開発2020

    • Author(s)
      村上 諒,庄野 逸,篠塚 寛志,永田 賢二,吉川 英樹
    • Organizer
      第81回日本応用物理学会秋季学術講演会
    • Related Report
      2020 Research-status Report
  • [Presentation] B-DCGAN: Evaluation of Binarized DCGAN for FPGA2019

    • Author(s)
      Hideo Terada, Hayaru Shouno
    • Organizer
      International Conference on Neural Information Processing (ICONIP2019),
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] Fast Bayesian Restoration of Poisson Corruputed Images with INLA2019

    • Author(s)
      Takahiro Kawashima, Hayaru Shouno
    • Organizer
      Int’l Conference on Parallel Distributed Processing Techniques and Applications(PDPTA) 2019
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] Implementation of an FPGA-based energy-efficient MCMC method for 2D Lenz-Ising model2019

    • Author(s)
      Patrick Tchicali, Hayaru Shouno
    • Organizer
      電子情報通信学会ニューロコンピューティング研究会
    • Related Report
      2019 Research-status Report

URL: 

Published: 2019-04-18   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi