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

2018 Fiscal Year Annual Research Report

New developments for big data by non-sparse modeling

Research Project

Project/Area Number 17K19956
Research InstitutionUniversity of Tsukuba

Principal Investigator

青嶋 誠  筑波大学, 数理物質系, 教授 (90246679)

Co-Investigator(Kenkyū-buntansha) 矢田 和善  筑波大学, 数理物質系, 准教授 (90585803)
石井 晶  東京理科大学, 理工学部情報科学科, 助教 (20801161)
赤平 昌文  筑波大学, 数理物質系(名誉教授), 名誉教授 (70017424)
Project Period (FY) 2017-06-30 – 2019-03-31
Keywords非スパースモデリング / スパイクノイズ / ビッグデータ / 人工知能 / ディープラーニング
Outline of Annual Research Achievements

本研究課題の最終年度として、非スパースモデリング技法の確立に取り組んだ。青嶋と矢田と赤平は、前年度までの研究によって、ビッグデータの潜在構造とノイズ構造について、非スパース性の評価基準を与え、非スパースなノイズ構造をスパース化するデータ変換法を開発した。青嶋と矢田は、一般に、ビッグデータのノイズが非スパース構造をもつ場合、このデータ変換による前処理を行わないと、潜在構造分析における様々な統計的推測に漸近正規性が成立しないことを証明した。データ変換を施すことで、非スパースノイズが除去されて潜在情報が浮き彫りになり、潜在構造の非スパース性を利用した非スパースモデリング技法を確立するに至った。さらに、青嶋と矢田と石井は、非スパースなノイズがある特殊な構造をもつ場合には、潜在構造とノイズ構造を同時に解析することで、データ変換よりも優れた性能をもつ非スパースモデリング技法が構築できることを示した。非スパースモデリング技法は、高精度かつ高速で汎用性が非常に高く、特徴量の抽出にも高速化が期待できる。本研究の成果は世界的に注目され、多数の招待講演を行った。特に、青嶋は台湾Academia Sinicaで開催された国際学会で基調講演を行い、矢田はハンガリーで開催されたIWAPの国際学会で招待講演、石井は日本数学会年会で特別講演を行った。さらに、本研究の成果の一部分は、青嶋・矢田による著書「高次元の統計学」(共立出版)の一部分になっている。

  • Research Products

    (23 results)

All 2019 2018 Other

All Int'l Joint Research (3 results) Journal Article (6 results) (of which Peer Reviewed: 6 results,  Open Access: 6 results) Presentation (11 results) (of which Int'l Joint Research: 10 results,  Invited: 11 results) Book (1 results) Remarks (1 results) Funded Workshop (1 results)

  • [Int'l Joint Research] Princeton University(米国)

    • Country Name
      U.S.A.
    • Counterpart Institution
      Princeton University
  • [Int'l Joint Research] Academia Sinica(中国)

    • Country Name
      CHINA
    • Counterpart Institution
      Academia Sinica
  • [Int'l Joint Research] Seoul National University(韓国)

    • Country Name
      KOREA (REP. OF KOREA)
    • Counterpart Institution
      Seoul National University
  • [Journal Article] Equality tests of high-dimensional covariance matrices under the strongly spiked eigenvalue model2019

    • Author(s)
      Ishii Aki、Yata Kazuyoshi、Aoshima Makoto
    • Journal Title

      Journal of Statistical Planning and Inference

      Volume: 202 Pages: 99~111

    • DOI

      10.1016/j.jspi.2019.02.002

    • Peer Reviewed / Open Access
  • [Journal Article] 日本統計学会賞受賞者特別寄稿論文:高次元統計解析: 理論と方法論の新しい展開2018

    • Author(s)
      青嶋 誠
    • Journal Title

      日本統計学会誌

      Volume: 48 Pages: 89~111

    • Peer Reviewed / Open Access
  • [Journal Article] High-Dimensional Quadratic Classifiers in Non-sparse Settings2018

    • Author(s)
      Aoshima Makoto、Yata Kazuyoshi
    • Journal Title

      Methodology and Computing in Applied Probability

      Volume: - Pages: -

    • DOI

      10.1007/s11009-018-9646-z

    • Peer Reviewed / Open Access
  • [Journal Article] Distance-based classifier by data transformation for high-dimension, strongly spiked eigenvalue models2018

    • Author(s)
      Aoshima Makoto、Yata Kazuyoshi
    • Journal Title

      Annals of the Institute of Statistical Mathematics

      Volume: - Pages: -

    • DOI

      10.1007/s10463-018-0655-z

    • Peer Reviewed / Open Access
  • [Journal Article] Inference on high-dimensional mean vectors under the strongly spiked eigenvalue model2018

    • Author(s)
      Ishii Aki、Yata Kazuyoshi、Aoshima Makoto
    • Journal Title

      Japanese Journal of Statistics and Data Science

      Volume: - Pages: -

    • DOI

      10.1007/s42081-018-0029-z

    • Peer Reviewed / Open Access
  • [Journal Article] 統計的推測理論の深化と進展のヒストリー2018

    • Author(s)
      赤平 昌文
    • Journal Title

      日本統計学会誌

      Volume: 47 Pages: 51~76

    • Peer Reviewed / Open Access
  • [Presentation] Non-Sparse Modeling for High-Dimensional Data2019

    • Author(s)
      Aoshima Makoto
    • Organizer
      Waseda International Symposium ``Introduction of General Causality to Various Data & its Applications"
    • Int'l Joint Research / Invited
  • [Presentation] A high-dimensional quadratic classifier under the strongly spiked eigenvalue model2019

    • Author(s)
      Yata Kazuyoshi、Ishii Aki、Aoshima Makoto
    • Organizer
      The 14th Workshop on Stochastic Models, Statistics and their Application
    • Int'l Joint Research / Invited
  • [Presentation] 強スパイク固有値モデルにおける高次元統計的推測2019

    • Author(s)
      石井晶
    • Organizer
      日本数学会2019年度年会
    • Invited
  • [Presentation] Tests of High-Dimensional Mean Vectors and Its Application Under the SSE Model2019

    • Author(s)
      Aki Ishii、Yata Kazuyoshi、Aoshima Makoto
    • Organizer
      Waseda International Symposium ``Introduction of General Causality to Various Data & its Applications"
    • Int'l Joint Research / Invited
  • [Presentation] High-Dimensional Statistical Analysis: Non-Sparse Modeling, Geometric Representations and New PCAs2018

    • Author(s)
      Aoshima Makoto
    • Organizer
      2018 Workshop on High-Dimensional Statistical Analysis
    • Int'l Joint Research / Invited
  • [Presentation] New techniques in high-dimensional statistical analysis2018

    • Author(s)
      Aoshima Makoto
    • Organizer
      Waseda International Symposium ``Introduction of General Causality to Various Data & Its Innovation of The Optimal Inference"
    • Int'l Joint Research / Invited
  • [Presentation] High-dimensional statistical analysis under spiked models2018

    • Author(s)
      Aoshima Makoto
    • Organizer
      The Fourth Conference of the International Society for Nonparametric Statistics
    • Int'l Joint Research / Invited
  • [Presentation] Inference on high-dimensional mean vectors under the strongly spiked eigenvalue model2018

    • Author(s)
      Yata Kazuyoshi、Aoshima Makoto
    • Organizer
      The Ninth International Workshop on Applied Probability
    • Int'l Joint Research / Invited
  • [Presentation] Tests of high-dimensional mean vectors under the SSE model2018

    • Author(s)
      Aki Ishii、Yata Kazuyoshi、Aoshima Makoto
    • Organizer
      International Symposium on Statistical Theory and Methodology for Large Complex Data
    • Int'l Joint Research / Invited
  • [Presentation] Equality tests of high-dimensional covariance matrices on the basis of strongly spiked eigenvalues2018

    • Author(s)
      Aki Ishii、Yata Kazuyoshi、Aoshima Makoto
    • Organizer
      Waseda International Symposium ``Introduction of General Causality to Various Data & Its Innovation of The Optimal Inference"
    • Int'l Joint Research / Invited
  • [Presentation] Equality tests for high-dimensional covariance matrices2018

    • Author(s)
      Aki Ishii、Yata Kazuyoshi、Aoshima Makoto
    • Organizer
      The 27th South Taiwan Statistics Conference
    • Int'l Joint Research / Invited
  • [Book] 高次元の統計学2019

    • Author(s)
      青嶋 誠、矢田 和善
    • Total Pages
      120
    • Publisher
      共立出版
    • ISBN
      978-4-320-11263-6
  • [Remarks] 青嶋研究室ホームページ

    • URL

      http://www.math.tsukuba.ac.jp/~aoshima-lab/jp/

  • [Funded Workshop] International Symposium on Statistical Theory and Methodology for Large Complex Data2018

URL: 

Published: 2019-12-27  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi