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

Development of efficient multi-armed bandit algorithm for good arm identification and its application

Research Project

Project/Area Number 18K18099
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 61030:Intelligent informatics-related
Research InstitutionHokkaido University

Principal Investigator

Koji Tabata  北海道大学, 電子科学研究所, 准教授 (20814445)

Project Period (FY) 2018-04-01 – 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: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2020: ¥390,000 (Direct Cost: ¥300,000、Indirect Cost: ¥90,000)
Fiscal Year 2019: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2018: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Keywords多腕バンディット / 良腕識別 / 分類バンディット / アルゴリズム / 機械学習 / 多腕バンディット問題 / 良腕識別問題
Outline of Final Research Achievements

The purpose of this research is development and application of efficient methods to identify good arms, whose expected reward is larger then a given threshold, under the multi-armed bandit setting which is a model of trade-off between knowledge exploration and exploitation. Here, an efficient method means that it can identify the good arm as few samples as possible.
We confirmed that our proposed methods have better performance then exiting method. We have also developed a prototype diagnostic device using the algorithm developed in this research.

Academic Significance and Societal Importance of the Research Achievements

学術的意義としては、病理診断に応用することを考慮した新しい問題設定とそのための良腕探索アルゴリズムの開発を行った。本研究により開発された手法は、病理診断だけではなく、品質保証、品質管理、創薬などへも応用可能である。
また、情報計測の分野にとって避けられない偽陽性・偽陰性が扱えるように、多腕バンディットアルゴリズムの平均報酬にバイアスが存在するような設定を新規に開拓した。

Report

(5 results)
  • 2021 Annual Research Report   Final Research Report ( PDF )
  • 2020 Research-status Report
  • 2019 Research-status Report
  • 2018 Research-status Report
  • Research Products

    (8 results)

All 2021 2019 2018

All Journal Article (2 results) (of which Peer Reviewed: 2 results,  Open Access: 2 results) Presentation (6 results) (of which Int'l Joint Research: 3 results)

  • [Journal Article] Classification Bandits: Classification Using Expected Rewards as Imperfect Discriminators2021

    • Author(s)
      Tabata Koji、Nakumura Atsuyoshi、Komatsuzaki Tamiki
    • Journal Title

      Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), Workshop on Machine Learning for MEasurement INformatics

      Volume: - Pages: 57-69

    • DOI

      10.1007/978-3-030-75015-2_6

    • ISBN
      9783030750145, 9783030750152
    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] A bad arm existence checking problem: How to utilize asymmetric problem structure?2019

    • Author(s)
      Tabata Koji、Nakamura Atsuyoshi、Honda Junya、Komatsuzaki Tamiki
    • Journal Title

      Machine Learning

      Volume: 109 Issue: 2 Pages: 327-372

    • DOI

      10.1007/s10994-019-05854-7

    • Related Report
      2019 Research-status Report
    • Peer Reviewed / Open Access
  • [Presentation] Classification Bandits: Classification Using Expected Rewards as Imperfect Discriminators.2021

    • Author(s)
      Koji Tabata
    • Organizer
      PAKDD (Workshops)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Classification Bandits: Classification Using Expected Rewards as Imperfect Discriminators2021

    • Author(s)
      Koji Tabata
    • Organizer
      PAKDD2021 Workshop on Machine Learning for MEasurement INformatics
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Bad Arm Existence Checking Algorithm with Bandit Feedback2019

    • Author(s)
      Koji Tabata, Atsuyoshi Nakamura, Tamiki Komatsuzaki
    • Organizer
      The 20th RIES-HOKUDAI International Symposium
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] 非対称なグレイゾーン付き悪腕存在チェックアルゴリズム2019

    • Author(s)
      田畑 公次, 中村 篤祥, 小松崎 民樹
    • Organizer
      離散構造処理系プロジェクト「2019年度 初夏のワークショップ」
    • Related Report
      2019 Research-status Report
  • [Presentation] バンディットフィードバック下での良腕悪腕の個数による分類問題2019

    • Author(s)
      田畑 公次, 中村 篤祥, 小松崎 民樹
    • Organizer
      第22回情報論的学習理論ワークショップ (IBIS 2019)
    • Related Report
      2019 Research-status Report
  • [Presentation] 非対称性を利用した悪腕存在チェックアルゴリズム2018

    • Author(s)
      田畑 公次, 中村 篤祥, 小松崎 民樹
    • Organizer
      平成30年度 情報論的学習理論と機械学習研究会
    • Related Report
      2018 Research-status Report

URL: 

Published: 2018-04-23   Modified: 2023-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi