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

Information geometrical hierarchical modeling

Research Project

Project/Area Number 20K19865
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 61030:Intelligent informatics-related
Research InstitutionKyushu Institute of Technology

Principal Investigator

Ishibashi Hideaki  九州工業大学, 大学院生命体工学研究科, 助教 (30838389)

Project Period (FY) 2020-04-01 – 2022-03-31
Project Status Completed (Fiscal Year 2021)
Budget Amount *help
¥2,990,000 (Direct Cost: ¥2,300,000、Indirect Cost: ¥690,000)
Fiscal Year 2021: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2020: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Keywords情報幾何学 / 階層モデリング / メタ学習 / マルチタスク学習 / ガウス過程 / 転移学習
Outline of Research at the Start

データから得られた知識を他の知識を学習する際に利用する枠組みはマルチタスク学習や転移学習,メタ学習と呼ばれる.既存のマルチタスク学習や転移学習,メタ学習の多くは学習させたいタスクに合わせてアルゴリズムが提案されており,汎用的に利用できる枠組みは少なかった.本研究では任意の学習タスクの幾何学構造を包括的に定義することで目的に合わせてシームレスにアルゴリズムを提供するための枠組みの構築を目指す.

Outline of Final Research Achievements

The purpose of this study is to develop a theory of hierarchical modeling for Bayes posteriors based on information geometry. For this purpose, we addressed the following three themes.(1)To define the structure of the set of Bayes posteriors having infinite-dimensional model parameter. (2)Development of a manifold modeling method for a set of Bayes posteriors based on kernel smoother. (3) Development of a hierarchical modeling method for a set of Bayes posteriors with latent variables.

Academic Significance and Societal Importance of the Research Achievements

本研究では情報幾何学に基づいて汎用的に利用できるメタ学習,マルチタスク学習,転移学習の方法論を構築した. 特に本研究の枠組みでは一般的に扱われる教師あり学習のメタ学習やマルチタスク学習だけでなく教師なし学習のメタ学習やマルチタスク学習も統一的に扱うことが可能となる. これにより様々なデータの形式や学習タスクに適した学習アルゴリズムをシームレスに提供できるようになった.

Report

(3 results)
  • 2021 Annual Research Report   Final Research Report ( PDF )
  • 2020 Research-status Report
  • Research Products

    (7 results)

All 2022

All Journal Article (3 results) (of which Peer Reviewed: 3 results,  Open Access: 2 results) Presentation (4 results) (of which Int'l Joint Research: 1 results)

  • [Journal Article] Principal Component Analysis for Gaussian Process Posteriors2022

    • Author(s)
      Ishibashi Hideaki、Akaho Shotaro
    • Journal Title

      Neural Computation

      Volume: 34 Issue: 5 Pages: 1189-1219

    • DOI

      10.1162/neco_a_01489

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [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 Annual Research Report
    • Peer Reviewed / Open Access
  • [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 Annual Research Report
    • Peer Reviewed / Open Access
  • [Presentation] Scalable manifold modeling by Nadaraya-Watson kernel regression2022

    • Author(s)
      Miyazaki Kazuki, Takano Shuhei, Tsuno Ryo, Ishibashi Hideaki, Furukawa Tetsuo
    • Organizer
      The 15th International Conference on Innovative Computing, Information and Control
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Sparse approximation of unsupervised kernel regressionfor large scale relational data2022

    • Author(s)
      Miyazaki Kazuki and Ishibashi Hideaki and Furukawa Tetsuo
    • Organizer
      The 3rd International Symposium on Neuromorphic AI Hardware
    • Related Report
      2021 Annual Research Report
  • [Presentation] Meta-modeling of manifold models for dynamical systems through biased optimal transport distance minimization2022

    • Author(s)
      Nakashima Seitaro, Ishibashi Hideaki, Furukawa Tetsuo
    • Organizer
      The 3rd International Symposium on Neuromorphic AI Hardware
    • Related Report
      2021 Annual Research Report
  • [Presentation] Simultaneous Meta-modeling of Dynamics and Kinematics based on the Hierarchical Manifold Modeling2022

    • Author(s)
      Tanka Daiki, Ishibashi Hideaki, Furukawa Tetsuo
    • Organizer
      The 3rd International Symposium on Neuromorphic AI Hardware
    • Related Report
      2021 Annual Research Report

URL: 

Published: 2020-04-28   Modified: 2023-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi