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

Generative manifold modeling of set of datasets based on optimal transport distance

Research Project

Project/Area Number 21K12061
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 InstitutionKyushu Institute of Technology

Principal Investigator

Furukawa Tetsuo  九州工業大学, 大学院生命体工学研究科, 教授 (50219101)

Co-Investigator(Kenkyū-buntansha) 石橋 英朗  九州工業大学, 大学院生命体工学研究科, 助教 (30838389)
Project Period (FY) 2021-04-01 – 2025-03-31
Project Status Completed (Fiscal Year 2024)
Budget Amount *help
¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000)
Fiscal Year 2023: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2022: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2021: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Keywordsメタモデリング / 最適輸送距離 / メタ学習 / マルチタスク学習 / マルチダイナミクス学習 / 直積潜在空間モデル / マルチダイナミクス / メタモデル / マルチドメイン / マルチビュー / 多様体モデリング / マルチレベルモデリング / データ集合体
Outline of Research at the Start

本研究の目的は,データ集合の集合(データ集合体)をモデリングする学習の理論と手法開発,およびその応用である.本研究ではデータ集合体を多様体を用いて確率モデル化するとともに,それらを確率モデル空間におけるリーマン多様体として階層的にモデル化する.その際,確率モデル間の距離を最適輸送距離で定義する.すなわち測地的最適輸送距離による階層的多様体モデリングの開発と学習理論究明が本研究の目標である.

Outline of Final Research Achievements

The proposed framework was applied to several domains, including the learning of multi-dynamics models of diverse walking patterns and the generation of novel movement patterns; the modeling of multi-genre item spaces for serendipity-enhancing recommendations; and the analysis of team tactics using multi-team modeling. These applications demonstrated the versatility and effectiveness of the developed approach.
The proposed framework was applied to several domains, including the learning of multi-dynamics models of diverse walking patterns and the generation of novel movement patterns; the modeling of multi-genre item spaces for serendipity-enhancing recommendations; and the analysis of team tactics using multi-team modeling. These applications demonstrated the versatility and effectiveness of the developed approach.

Academic Significance and Societal Importance of the Research Achievements

本研究は、多様なタスク群の共通構造を抽出するメタモデリングに対し、最適輸送距離(OT)に基づく新たな学習理論を提案した点に学術的意義がある。観測空間と直積潜在空間の間にOT写像を定義し、その最適化により教師なし・小標本でも頑健なメタ学習が可能となる枠組みを構築した。また、情報幾何学との関連も明確にし、OTの理論的位置づけを与えた。
社会的には、本手法は「多様なモデルの集合」で表される現象に広く適用できる。歩行や運動の多様性、音楽ジャンル、チーム戦術など、個別モデルでは扱いにくい構造の理解と新たな創発の支援が可能となる。

Report

(5 results)
  • 2024 Annual Research Report   Final Research Report ( PDF )
  • 2023 Research-status Report
  • 2022 Research-status Report
  • 2021 Research-status Report
  • Research Products

    (24 results)

All 2025 2024 2023 2022 2021

All Journal Article (3 results) (of which Peer Reviewed: 3 results,  Open Access: 2 results) Presentation (21 results) (of which Int'l Joint Research: 15 results)

  • [Journal Article] Low-rank kernel decomposition for scalable manifold modeling2022

    • Author(s)
      K. Miyazaki, S. Takano, R. Tsuno, H. Ishibashi, T. Furukawa
    • Journal Title

      2022 Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2022

      Volume: - Pages: 1-6

    • DOI

      10.1109/scisisis55246.2022.10001865

    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] Visual analytics of set data for knowledge discovery and member selection support2022

    • Author(s)
      Watanabe Ryuji、Ishibashi Hideaki、Furukawa Tetsuo
    • Journal Title

      Decision Support Systems

      Volume: 152 Pages: 113635-113635

    • DOI

      10.1016/j.dss.2021.113635

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Multi-task manifold learning for small sample size datasets2022

    • Author(s)
      Ishibashi Hideaki、Higa Kazushi、Furukawa Tetsuo
    • Journal Title

      Neurocomputing

      Volume: 473 Pages: 138-157

    • DOI

      10.1016/j.neucom.2021.11.043

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Presentation] Visual exploration system for discovering favorite Arita porcelain types2025

    • Author(s)
      Daichi Yoshikane, Hideaki Ishibashi, Tetsuo Furukawa,
    • Organizer
      The 6th International Symposium on Neuromorphic AI Hardware
    • Related Report
      2024 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Unsupervised Meta-Modeling of a Set of Dynamical Systems: Latent Variable Estimation in State and Observation Spaces2025

    • Author(s)
      Yuki Tokunaga, Hideaki Ishibashi, Tetsuo Furukawa
    • Organizer
      The 6th International Symposium on Neuromorphic AI Hardware
    • Related Report
      2024 Annual Research Report
    • Int'l Joint Research
  • [Presentation] On the visualization of contextual meaning in waka2025

    • Author(s)
      Soichiro Takemura, Hideaki Ishibashi, Tetsuo Furukawa
    • Organizer
      The 6th International Symposium on Neuromorphic AI Hardware
    • Related Report
      2024 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Viewpoint Representation via Dual-Space Embedding of Object-Tag Data2025

    • Author(s)
      Taisei Sakaguchi, Hideaki Ishibashi, Tetsuo Furukawa
    • Organizer
      The 6th International Symposium on Neuromorphic AI Hardware
    • Related Report
      2024 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Meta-Modeling of Latent Variable Models Based on the Optimal Transport Distance2025

    • Author(s)
      Tetsuo Furukawa, Hideaki Ishibashi
    • Organizer
      The 6th International Symposium on Neuromorphic AI Hardware
    • Related Report
      2024 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Multi-task Gaussian process Based on the Combination of Two Types of Dimensionality Reduction Techniques2024

    • Author(s)
      Hideaki Ishibashi
    • Organizer
      12th International Symposium on Applied Engineering and Sciences (SAES2024)
    • Related Report
      2024 Annual Research Report
    • Int'l Joint Research
  • [Presentation] セレンディピティを促進する情報探索システムのデザインと実現2024

    • Author(s)
      金津 達也,吉兼 大智,石橋 英朗,吉田 香,古川 徹生
    • Organizer
      日本神経回路学会
    • Related Report
      2024 Annual Research Report
  • [Presentation] 視点を表現する Object-Tag データの双対埋め込み2024

    • Author(s)
      広渡 朱莉,坂口 大征,石橋 英朗,古川 徹生
    • Organizer
      日本神経回路学会
    • Related Report
      2024 Annual Research Report
  • [Presentation] 情報検索・探索システムにおけるセレンディピティ2024

    • Author(s)
      金津 達也,吉田 香,古川 徹生
    • Organizer
      日本認知科学会
    • Related Report
      2024 Annual Research Report
  • [Presentation] 潜在変数モデルのメタモデリング:最適輸送距離の観点による考察2024

    • Author(s)
      古川 徹生, 石橋 英朗
    • Organizer
      日本人工知能学会
    • Related Report
      2024 Annual Research Report
  • [Presentation] Connotation Visualization of ‘uta-kotoba’ of Waka poetry Based on Conditional Co-occurrence Probability2024

    • Author(s)
      Yusuke Ayukawa, Tetsuo Furukawa
    • Organizer
      The 5th International Symposium on Neuromorphic AI Hardware
    • Related Report
      2023 Research-status Report
    • Int'l Joint Research
  • [Presentation] Development of an information retrieval and exploration system for identifying user's domain of interest2024

    • Author(s)
      Ryusei Iki, Tetsuo Furukawa
    • Organizer
      The 5th International Symposium on Neuromorphic AI Hardware
    • Related Report
      2023 Research-status Report
    • Int'l Joint Research
  • [Presentation] ダイナミクスとキネマティクスのメタモデリング:最適輸送最小化によるアプローチ2023

    • Author(s)
      德永優希,竹村綜一朗,田中大揮,石橋英朗,古川徹生
    • Organizer
      日本神経回路学会全国大会
    • Related Report
      2023 Research-status Report
  • [Presentation] 相互情報量最大化に基づくヘテロ共起データの埋め込み法の提案2023

    • Author(s)
      石田 琢朗, 古川 徹生
    • Organizer
      日本神経回路学会全国大会
    • Related Report
      2023 Research-status Report
  • [Presentation] Co-training Based Metric Learning for Manifold Modeling2023

    • Author(s)
      Yuki Fukunaga,Shuri hirowatari,Tetsho Furukawa
    • Organizer
      International Conference on Innovative Computing, Information and Control
    • Related Report
      2023 Research-status Report
    • Int'l Joint Research
  • [Presentation] Heterogeneous Co-occurrence Embedding via Mutual Information Maximization2023

    • Author(s)
      Takuro Ishida, Keisuke Yoneda,Tetsuo Furukawa
    • Organizer
      International Symposium on Advanced Intelligent Systems
    • Related Report
      2023 Research-status Report
    • Int'l Joint Research
  • [Presentation] Meta-modeling of manifold models for dynamical systems through biased optimal transport distance minimization2022

    • Author(s)
      S. Nakashima, H. Ishibashi, T. Furukawa
    • Organizer
      The 3rd International Symposium on Neuromorphic AI Hardware
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Simultaneous Meta- modeling of Dynamics and Kinematics based on the Hierarchical Manifold Modeling2022

    • Author(s)
      D. Tanaka, H. Ishibashi, T. Furukawa
    • Organizer
      The 3rd International Symposium on Neuromorphic AI Hardware
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Sparse approximation of unsupervised kernel regressionfor large scale relational data2022

    • Author(s)
      K. Miyazaki, H. Ishibashi, T. Furukawa
    • Organizer
      The 3rd International Symposium on Neuromorphic AI Hardware
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Fashion outfit retrieval via hashtag search and visually-assisted browsing on jointed manifold models2021

    • Author(s)
      S. Hirowatari, T. Ishida, T. Iwasaki, T. Furukawa
    • Organizer
      International Conference on Innovative Computing, Information and Control, 2021
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Scalable manifold modeling by Nadaraya-Watson kernel regression2021

    • Author(s)
      K. Miyazaki, H. Ishibashi, T. Furukawa
    • Organizer
      International Conference on Innovative Computing, Information and Control, 2021
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research

URL: 

Published: 2021-04-28   Modified: 2026-01-16  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi