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

Research and Development of Experiment Candidate Recommendation System with Reasoning and Learning

Research Project

Project/Area Number 16K00304
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Intelligent informatics
Research InstitutionKyushu University

Principal Investigator

Koshimura Miyuki  九州大学, システム情報科学研究院, 助教 (30274492)

Co-Investigator(Kenkyū-buntansha) 力 規晃  徳山工業高等専門学校, 情報電子工学科, 助教 (50290804)
藤田 博  九州大学, システム情報科学研究院, 准教授 (70284552)
Research Collaborator Uemura Naoki  
Zha Aolong  
Asakura Kenta  
Project Period (FY) 2016-04-01 – 2019-03-31
Project Status Completed (Fiscal Year 2018)
Budget Amount *help
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2018: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2017: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2016: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Keywords機械学習応用 / 組合せ最適化 / SAT / 深層学習 / 顕微鏡画像の分類 / SAT符号化 / 最適化問題 / 機械学習 / 顕微鏡画像 / 電流の時系列データ / 帰納論理プログラミング / データマイニング / 顕微鏡画像の解析 / 人工知能 / データ解析
Outline of Final Research Achievements

We engaged in "analyzing microscopic image of urine formed element" and "abnormal diagnosis based on current data" with techniques of machine learning. We cut out images of approximately 14,000 cells from photomicrographs of 100 urine samples collected at a hospital and classified them into red blood cells, white blood cells, and others. We succeeded in acquiring a model with a correct answer rate of 98.30 %.
We also worked on MaxSAT encoding of coalition structure generation (CSG) problem and ILP (inductive logic programming) as case studies of problem solving with SAT technology. And we succeeded in achieving an order of magnitude performance improvement than a previous method for CSG.

Academic Significance and Societal Importance of the Research Achievements

本研究は、目標性能達成のため、これまでの実験データを基に次に試行すべき実験候補を推薦するシステムの開発を目指した。統計的手法に論理的手法を組み合わせることにより、探索や説明機能を強化する。ILPをSAT符号化して高速化を図る点も研究の特色である。また、多くの問題に現れる推移律の効率的なSAT符号化法を考案しその効果を実証した点は、組合せ最適化解法の適用分野を広げる成果である。

Report

(4 results)
  • 2018 Annual Research Report   Final Research Report ( PDF )
  • 2017 Research-status Report
  • 2016 Research-status Report
  • Research Products

    (21 results)

All 2019 2018 2017 2016 Other

All Journal Article (5 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 5 results) Presentation (15 results) (of which Int'l Joint Research: 2 results,  Invited: 1 results) Remarks (1 results)

  • [Journal Article] N-level Modulo-Based CNF encodings of Pseudo-Boolean constraints for MaxSAT2019

    • Author(s)
      Aolong Zha, Miyuki Koshimura, Hiroshi Fujita
    • Journal Title

      Constraints

      Volume: 24 Issue: 2 Pages: 133-161

    • DOI

      10.1007/s10601-018-9299-0

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] 分割決定木を用いた分割関数ゲームの提携構造形成アルゴリズム2019

    • Author(s)
      査 澳龍、 越村 三幸、櫻井 祐子、横尾 真
    • Journal Title

      電子情報通信学会論文誌 D

      Volume: J102-D Pages: 313-323

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Improved WPM encoding for coalition structure generation under MC-nets2018

    • Author(s)
      Liao Xiaojuan、Koshimura Miyuki、Nomoto Kazuki、Ueda Suguru、Sakurai Yuko、Yokoo Makoto
    • Journal Title

      Constraints

      Volume: 24 Issue: 1 Pages: 25-55

    • DOI

      10.1007/s10601-018-9295-4

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] SATとラムゼー数 ~数学の未解決問題への挑戦~2016

    • Author(s)
      藤田 博、越村 三幸
    • Journal Title

      情報処理

      Volume: 57 Pages: 716-719

    • Related Report
      2016 Research-status Report
    • Peer Reviewed
  • [Journal Article] MaxSAT:SATの最適化問題への拡張 -MaxSATソルバーの活用法-2016

    • Author(s)
      越村 三幸、藤田 博
    • Journal Title

      情報処理

      Volume: 57 Pages: 730-733

    • Related Report
      2016 Research-status Report
    • Peer Reviewed
  • [Presentation] Coalition Structure Generation for Partition Function Games Utilizing a Concise Graphical Representation2019

    • Author(s)
      Miyuki Koshimura, Aolong Zha, Kazuki Nomoto, Suguru Ueda, Yuko Sakurai, Makoto Yokoo
    • Organizer
      The 4th Kakenhi Kiban-A&B/NII Collaborate Research Meeting
    • Related Report
      2018 Annual Research Report
  • [Presentation] Improved WPM Encoding for Coalition Structure Generation under MC-nets,2019

    • Author(s)
      Xiaojuan Liao, Miyuki Koshimura, Kazuki Nomoto, Suguru Ueda, Yuko Sakurai, Makoto Yokoo
    • Organizer
      The 4th Kakenhi Kiban-A&B/NII Collaborate Research Meeting
    • Related Report
      2018 Annual Research Report
  • [Presentation] SATソルバーを用いたMCS列挙の実装とその評価2019

    • Author(s)
      越村 三幸,佐藤 健
    • Organizer
      人工知能学会 第109回人工知能基本問題研究会
    • Related Report
      2018 Annual Research Report
  • [Presentation] CNNを用いた尿中有形成分の分類の初期検討2018

    • Author(s)
      淺倉 健太,越村 三幸,藤田 博
    • Organizer
      電子情報通信学会 医用画像研究会
    • Related Report
      2018 Annual Research Report
  • [Presentation] MC-netsを用いた提携構造形成問題のMaxSAT符号化の改良と評価2018

    • Author(s)
      越村 三幸,廖 暁鵑,野本 一貴,上田 俊、櫻井 祐子,横尾 真
    • Organizer
      人工知能学会 第106回人工知能基本問題研究会
    • Related Report
      2017 Research-status Report
  • [Presentation] 機械学習を用いた電流データに基づく異常判定システム2018

    • Author(s)
      満尾 成亮,安部 竜太,橋下 司,越村 三幸,藤田 博
    • Organizer
      電子情報通信学会 2018年総合大会
    • Related Report
      2017 Research-status Report
  • [Presentation] 分割関数ゲームを対象とした提携構造形成問題のMaxSAT符号化2017

    • Author(s)
      越村 三幸,査 澳龍,野本 一貴,櫻井 祐子,横尾 真
    • Organizer
      2017年度 人工知能学会全国大会(第31回)
    • Related Report
      2017 Research-status Report
  • [Presentation] MC-netsを用いた提携構造形成問題のMaxSAT符号化の改良2017

    • Author(s)
      越村 三幸,廖 暁鵑,野本 一貴,上田 俊、櫻井 祐子,横尾 真
    • Organizer
      日本ソフトウェア科学会 第34回大会
    • Related Report
      2017 Research-status Report
  • [Presentation] Coalition Structure Generation for Partition Function Games Utilizing a Concise Graphical Representation2017

    • Author(s)
      Aolong Zha, Kazuki Nomoto, Suguru Ueda, Miyuki Koshimura, Yuko Sakurai, Makoto Yokoo
    • Organizer
      PRIMA 2017: Principles and Practice of Multi-Agent Systems - 20th International Conference
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] Mixed Radix Weight Totalizer Encoding for Pseudo-Boolean Constraints2017

    • Author(s)
      Aolong Zha, Naoki Uemura, Miyuki Koshimura, and Hiroshi Fujita
    • Organizer
      ICTAI 2017: 29th International Conference on Tools with Artificial Intelligence
    • Related Report
      2017 Research-status Report
    • Int'l Joint Research
  • [Presentation] A Hybrid Encoding of Pseudo-Boolean Constraints into CNF2017

    • Author(s)
      Aolong Zha, Miyuki Koshimura, and Hiroshi Fujita
    • Organizer
      Conference on Technologies and Applications of Artificial Intelligence
    • Related Report
      2017 Research-status Report
  • [Presentation] 混合基数を用いた擬似ブール制約のSAT符号化2017

    • Author(s)
      上村 直輝,藤田 博,越村 三幸,査 澳龍
    • Organizer
      人工知能学会 第103回人工知能基本問題研究会(SIG-FPAI)
    • Place of Presentation
      湯布院公民館
    • Related Report
      2016 Research-status Report
  • [Presentation] 分割決定木で表現された提携構造形成問題のMaxSAT符号化2017

    • Author(s)
      越村 三幸,査 澳龍,野本 一貴,櫻井 祐子,横尾 真
    • Organizer
      情報処理学会第79回全国大会
    • Place of Presentation
      名古屋大学東山キャンパス
    • Related Report
      2016 Research-status Report
  • [Presentation] モデルとUnsat Coreを利用したMaxSATソルバーの試作2016

    • Author(s)
      越村 三幸
    • Organizer
      2016年度人工知能学会全国大会(第30回)
    • Place of Presentation
      北九州国際会議場
    • Related Report
      2016 Research-status Report
  • [Presentation] MaxSAT Encoding for MC-net/PDT-based Coalition Structure Generation Problem2016

    • Author(s)
      Miyuki Koshimura, Kazuki Nomoto, Yuko Sakurai, and Makoto Yokoo
    • Organizer
      10th CSPSAT Seminar
    • Place of Presentation
      神戸大学六甲台キャンパス
    • Related Report
      2016 Research-status Report
    • Invited
  • [Remarks] QMaxSAT: Q-dai MaxSAT Solver

    • URL

      https://sites.google.com/site/qmaxsat/

    • Related Report
      2016 Research-status Report

URL: 

Published: 2016-04-21   Modified: 2020-03-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi