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

2023 Fiscal Year Annual Research Report

Innovative Developments of Theories and Methodologies for Large Complex Data

Research Project

Project/Area Number 20H00576
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, 教授 (90585803)
荒木 由布子  東北大学, 情報科学研究科, 教授 (80403913)
川野 秀一  九州大学, 数理学研究院, 教授 (50611448)
蛭川 潤一  新潟大学, 自然科学系, 准教授 (10386617)
鈴木 大慈  東京大学, 大学院情報理工学系研究科, 准教授 (60551372)
石井 晶  東京理科大学, 創域理工学部情報計算科学科, 講師 (20801161)
Project Period (FY) 2020-04-01 – 2025-03-31
Keywords高次元データ / 時空間データ / 高次元統計解析 / 深層学習 / 高次元小標本
Outline of Annual Research Achievements

青嶋と矢田は、自動正則化主成分分析法とデータ変換法を考案し、高次元ノイズの高精度な除去を可能にした。高次元非階層型クラスタリングが一致性をもつ条件を示した。 自動正則化主成分分析法を使いデータの非スパース構造を解析して汚染と正常を精密に識別し、データ変換法を使い汚染データの正常化に成功した。 青嶋と矢田と石井は、深層学習の中間層に潜む非スパースな構造に汎化誤差の精密な評価をした。川野は、マルチタスク学習モデルにおいて、凸クラスタリングをモデリングの過程に組み込み、推定精度および予測精度を改善した。ベイズスパースモデルについて、連結lassoを馬蹄事前分布の枠組みに拡張した。蛭川は、独立性の帰無仮説に対して、定常ARMA対立仮説を検定する問題を局所定常過程に一般化し、ランク検定統計量を応用した。Le Camのcontiguityの概念とLe Cam’s third lemmaを応用し、帰無仮説と対立仮説の両方で検定統計量の漸近正規性を導いた。鈴木は、高次元データにおけるニューラルネットワークの学習に最適化と予測誤差の解析を行った。平均場ニューラルネットワークの学習について、収束解析と次元に対するサンプル複雑度を導出し、カーネル法に優越することを理論的に示した。データの非等方性が計算量と統計的予測誤差の両方を改善し、カーネル法よりも強く恩恵を受けることを示した。深層基盤モデルに関して近似誤差と汎化誤差を解析し、拡散モデルとTransformerが次元の呪いを回避する機構を備えていることを示した。荒木は、長期大規模コホート研究における高精度な生存時間推定法の開発に取り組み、時間依存共変量がある場合に生存関数推定のバイアスを除くモデルを検討した。
得られた研究成果は国内外の学会や学術誌で発表し、さらに、研究テーマに沿ったシンポジウムを筑波大学・新潟大学・九州大学・東北大学で開催した。

Current Status of Research Progress
Current Status of Research Progress

2: Research has progressed on the whole more than it was originally planned.

Reason

研究計画の4年目として、多様な大規模複雑データを解析するための理論と方法論が、盤石に構築されてきた。特に、
(2)汚染データに頑健な非スパースモデリングの構築
(4)高次元統計解析による深層学習の解明
(5)大規模時空間データへの非スパースモデリングの拡張
については躍進的な成果が得られ、おおむね順調に進展していると評価できる。

Strategy for Future Research Activity

研究課題の最終年度では、研究目的に掲げた5点について、理論と方法論の研究成果を盤石なものにする。また、研究課題について総括的な国際シンポジウムを開催し、科学技術・社会・経済・産業の要請に多大な貢献をもたらすことを目指す。

  • Research Products

    (42 results)

All 2025 2024 2023 Other

All Int'l Joint Research (6 results) Journal Article (11 results) (of which Int'l Joint Research: 6 results,  Peer Reviewed: 11 results,  Open Access: 10 results) Presentation (21 results) (of which Int'l Joint Research: 17 results,  Invited: 18 results) Book (1 results) Remarks (2 results) Funded Workshop (1 results)

  • [Int'l Joint Research] Princeton University/University of North Carolina/University of California, Davis(米国)

    • Country Name
      U.S.A.
    • Counterpart Institution
      Princeton University/University of North Carolina/University of California, Davis
    • # of Other Institutions
      2
  • [Int'l Joint Research] Academia Sinica/National Taiwan University(中国)

    • Country Name
      CHINA
    • Counterpart Institution
      Academia Sinica/National Taiwan University
  • [Int'l Joint Research] Seoul National University/韓国科学技術院(韓国)

    • Country Name
      KOREA (REP. OF KOREA)
    • Counterpart Institution
      Seoul National University/韓国科学技術院
  • [Int'l Joint Research] University of Stavanger(ノルウェー)

    • Country Name
      NORWAY
    • Counterpart Institution
      University of Stavanger
  • [Int'l Joint Research] National University of Singapore(シンガポール)

    • Country Name
      SINGAPORE
    • Counterpart Institution
      National University of Singapore
  • [Int'l Joint Research]

    • # of Other Countries
      2
  • [Journal Article] Automatic Sparse PCA for High-Dimensional Data2025

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

      Statistica Sinica

      Volume: 35 Pages: ー

    • DOI

      10.5705/ss.202022.0319

    • Peer Reviewed / Open Access
  • [Journal Article] Asymptotic properties of hierarchical clustering in high-dimensional settings2024

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

      Journal of Multivariate Analysis

      Volume: 199 Pages: 105251~105251

    • DOI

      10.1016/j.jmva.2023.105251

    • Peer Reviewed / Open Access
  • [Journal Article] High-dimensional Statistical Analysis and Its Application to an ALMA Map of NGC 2532024

    • Author(s)
      Takeuchi Tsutomu T.、Yata Kazuyoshi、Egashira Kento、Aoshima Makoto、Ishii Aki、Cooray Suchetha、Nakanishi Kouichiro、Kohno Kotaro、Kono Kai T.
    • Journal Title

      The Astrophysical Journal Supplement Series

      Volume: 271 Pages: 44~44

    • DOI

      10.3847/1538-4365/ad2517

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Multi-task learning regression via convex clustering2024

    • Author(s)
      Okazaki Akira、Kawano Shuichi
    • Journal Title

      Computational Statistics & Data Analysis

      Volume: 195 Pages: 107956~107956

    • DOI

      10.1016/j.csda.2024.107956

    • Peer Reviewed / Open Access
  • [Journal Article] Anisotropy helps: improved statistical and computational complexity of the mean-field Langevin dynamics under structured data2024

    • Author(s)
      Atsushi Nitanda, Kazusato Oko, Taiji Suzuki, Denny Wu
    • Journal Title

      The Twelfth International Conference on Learning Representations

      Volume: ー Pages: ー

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Bayesian fused lasso modeling via horseshoe prior2023

    • Author(s)
      Kakikawa Yuko、Shimamura Kaito、Kawano Shuichi
    • Journal Title

      Computational Statistics & Data Analysis

      Volume: 6 Pages: 705~727

    • DOI

      10.1007/s42081-023-00213-2

    • Peer Reviewed / Open Access
  • [Journal Article] Convergence of mean-field Langevin dynamics: Time and space discretization, stochastic gradient, and variance reduction2023

    • Author(s)
      Taiji Suzuki, Denny Wu, Atsushi Nitanda
    • Journal Title

      Advances in Neural Information Processing Systems

      Volume: 37 Pages: ー

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Feature learning via mean-field Langevin dynamics: classifying sparse parities and beyond2023

    • Author(s)
      Taiji Suzuki, Denny Wu, Kazusato Oko, Atsushi Nitanda
    • Journal Title

      Advances in Neural Information Processing Systems

      Volume: 37 Pages: ー

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Diffusion Models are Minimax Optimal Distribution Estimators2023

    • Author(s)
      Kazusato Oko, Shunta Akiyama, Taiji Suzuki
    • Journal Title

      Proceedings of the 40th International Conference on Machine Learning

      Volume: 202 Pages: 26517~26582

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Approximation and Estimation Ability of Transformers for Sequence-to-Sequence Functions with Infinite Dimensional Input2023

    • Author(s)
      Shokichi Takakura, Taiji Suzuki
    • Journal Title

      Proceedings of the 40th International Conference on Machine Learning

      Volume: 202 Pages: 33416~33447

    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Rank Tests for Randomness Against Time-Varying MA Alternative2023

    • Author(s)
      Hirukawa Junichi、Sakai Shunsuke
    • Journal Title

      In: Liu, Y., Hirukawa, J., Kakizawa, Y. (eds) Research Papers in Statistical Inference for Time Series and Related Models - Essays in Honor of Masanobu Taniguchi

      Volume: ー Pages: 221~245

    • DOI

      10.1007/978-981-99-0803-5_9

    • Peer Reviewed
  • [Presentation] Inference on high-dimensional mean vectors by the data transformation technique2024

    • Author(s)
      Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      IMS Asia Pacific Rim Meeting 2024
    • Int'l Joint Research / Invited
  • [Presentation] Asymptotic properties of kernel k-means under high dimensional settings2024

    • Author(s)
      Kento Egashira, Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      IMS Asia Pacific Rim Meeting 2024
    • Int'l Joint Research / Invited
  • [Presentation] Convergence of mean field Langevin dynamics and its application to neural network optimization2024

    • Author(s)
      Taiji Suzuki
    • Organizer
      The Mathematics of Data
    • Int'l Joint Research / Invited
  • [Presentation] Non-sparse modeling for high-dimensional data2023

    • Author(s)
      Makoto Aoshima
    • Organizer
      Statistical Week 2023(基調講演)
    • Int'l Joint Research / Invited
  • [Presentation] Threshold-based PCA in high-dimensional settings2023

    • Author(s)
      Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      TMU International Conference on Statistical Modelling and Inference 2023
    • Int'l Joint Research / Invited
  • [Presentation] Threshold-based Sparse PCA for high-dimensional data based on the noise-reduction methodology2023

    • Author(s)
      Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      Statistical Week 2023
    • Int'l Joint Research / Invited
  • [Presentation] Automatic sparse PCA for high-dimensional data and its applications2023

    • Author(s)
      Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      Seminar on Institute of Statistical Science, Academia Sinica
    • Int'l Joint Research / Invited
  • [Presentation] Estimation of the strongly spiked eigenstructure in high-dimensional settings2023

    • Author(s)
      Kazuyoshi Yata, Aki Ishii, Makoto Aoshima
    • Organizer
      The 6th International Conference on Econometrics and Statistics
    • Int'l Joint Research / Invited
  • [Presentation] Contrastive principal component analysis in high dimension low sample size2023

    • Author(s)
      Shao-Hsuan Wang, Kazuyoshi Yata
    • Organizer
      The 6th International Conference on Econometrics and Statistics
    • Int'l Joint Research / Invited
  • [Presentation] Quadratic classifiers for high-dimensional noisy data2023

    • Author(s)
      Aki Ishii, Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      The 6th International Conference on Econometrics and Statistics
    • Int'l Joint Research / Invited
  • [Presentation] Asymptotic behaviors of k-means under high dimensional settings2023

    • Author(s)
      Kento Egashira, Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      The 6th International Conference on Econometrics and Statistics
    • Int'l Joint Research / Invited
  • [Presentation] Asymptotic properties of kernel k-means for high dimensional data2023

    • Author(s)
      Kento Egashira, Kazuyoshi Yata, Makoto Aoshima
    • Organizer
      International Symposium on Recent Advances in Theories and Methodologies for Large Complex Data
    • Int'l Joint Research / Invited
  • [Presentation] 強スパイク固有値モデルにおける高次元共分散行列の統計的推測2023

    • Author(s)
      矢田和善, 石井 晶, 青嶋 誠
    • Organizer
      統計関連学会連合大会 応用統計学会企画セッション「高次元統計解析の最近の発展」
    • Invited
  • [Presentation] Automatic sparse PCA and its applications2023

    • Author(s)
      矢田和善
    • Organizer
      Seminar on Bayesian Computation
    • Invited
  • [Presentation] Functional mixture cure model and its application2023

    • Author(s)
      Yuko Araki
    • Organizer
      The 6th International Conference on Econometrics and Statistics
    • Int'l Joint Research / Invited
  • [Presentation] 関数データに基づく混合治癒モデルの提案2023

    • Author(s)
      荒木由布子
    • Organizer
      第37回日本計算機統計学会
  • [Presentation] 外れ値タスクに頑健な凸クラスタリング回帰2023

    • Author(s)
      岡崎彰良,川野秀一
    • Organizer
      2023年度統計関連学会連合大会
  • [Presentation] Multi-task learning regression based on convex clustering2023

    • Author(s)
      Okazaki, A. and Kawano, S.
    • Organizer
      The 16th International Conference of the ERCIM WG on Computational and Methodological Statistics
    • Int'l Joint Research
  • [Presentation] Feature learning via mean field neural networks and anisotropic features2023

    • Author(s)
      Taiji Suzuki, Denny Wu, Atsushi Nitanda and Kazusato Oko
    • Organizer
      International Symposium on Recent Advances in Theories and Methodologies for Large Complex Data
    • Int'l Joint Research / Invited
  • [Presentation] Rank tests for randomness against time-varying MA alternative2023

    • Author(s)
      Hirukawa, J.
    • Organizer
      The 6th International Conference on Econometrics and Statistics
    • Int'l Joint Research / Invited
  • [Presentation] Innovation algorithm of fractionally integrated (I(d)) process and applications on the estimation of parameters2023

    • Author(s)
      Hirukawa, J., Fujimori, K.
    • Organizer
      International Conference on Time Series Econometrics and Model Checking
    • Int'l Joint Research / Invited
  • [Book] Research Papers in Statistical Inference for Time Series and Related Models - Essays in Honor of Masanobu Taniguchi2023

    • Author(s)
      Liu, Y., Hirukawa, J., Kakizawa, Y.
    • Total Pages
      607
    • Publisher
      Springer
    • ISBN
      978-981-99-0802-8
  • [Remarks] 青嶋研究室ホームページ

    • URL

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

  • [Remarks] 青嶋研究室ホームページ 科研費基盤研究(A) シンポジウム

    • URL

      https://www.math.tsukuba.ac.jp/~aoshima-lab/jp/kiban_A.html

  • [Funded Workshop] International Symposium on Recent Advances in Theories and Methodologies for Large Complex Data2023

URL: 

Published: 2024-12-25  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi