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

Multi-agent Reinforcement Learning for Cooperative Policy with Different Abstraction

Research Project

Project/Area Number 20K23326
Research Category

Grant-in-Aid for Research Activity Start-up

Allocation TypeMulti-year Fund
Review Section 1001:Information science, computer engineering, and related fields
Research InstitutionOkayama University

Principal Investigator

Uwano Fumito  岡山大学, 自然科学学域, 助教 (30880687)

Project Period (FY) 2020-09-11 – 2022-03-31
Project Status Completed (Fiscal Year 2021)
Budget Amount *help
¥2,860,000 (Direct Cost: ¥2,200,000、Indirect Cost: ¥660,000)
Fiscal Year 2021: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2020: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Keywordsマルチエージェントシステム / 強化学習 / ニューラルネットワーク / 情報粒度 / 協調 / 抽象度
Outline of Research at the Start

本研究では,ロボットのように周囲の環境から得た情報を基に行動を決める主体(エージェント)が,複数集まったときの適切な行動則を獲得するマルチエージェント強化学習を実用化する上での,センサの個体差や状況の違いによる,観測情報の粒度の違いに適応した協調行動学習法を提案する.具体的には,エージェントにおける情報の抽象度を制御し,獲得情報の粒度に従ってエージェント毎の抽象度を調整することで,適切な協調行動を学習する.

Outline of Final Research Achievements

This research analyzed deep reinforcement learning agents’ performance in multiagent system with agents having different resolution in input each other to clarify the neural network can abstract the resolution appropriately. Furthermore, this research extended the previous method which enable agents to learn cooperative policy each other in dynamic environment into deep reinforcement learning to result the agents learned a cooperative policy in multiagent maze problem with agents having different resolution in input. At the end, this research introduced LSTM which can learn in time-sequential data into the proposed method to result that the agents can learn synchronously in that maze problem with environment being extended to dynamic one.

Academic Significance and Societal Importance of the Research Achievements

本研究により,従来のマルチエージェント強化学習では取り上げられることのなかった入力情報の粒度の異なる状況に対する追従という新たな学問分野を切り開くことができた.また,実問題に即して考えてみても,例えば複数ロボットの協働制御を考えたときに,ロボットごとのセンサの粒度が異なることや,故障などの状況により得られる情報の粒度が変化することは一般的だが,マルチエージェント強化学習ではあまり考えられることがなかったため,実問題における性能がシミュレーションと比べて高くない傾向にあった.本研究成果による方法論で,マルチエージェント強化学習を実問題に応用する上での性能向上に寄与できたと考えられる.

Report

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

    (6 results)

All 2022 2021

All Journal Article (1 results) (of which Peer Reviewed: 1 results,  Open Access: 1 results) Presentation (5 results) (of which Int'l Joint Research: 3 results,  Invited: 1 results)

  • [Journal Article] Sigmoid-based Incorrect Opinion Prevention Algorithm on Multi-Opinion Sharing Model2021

    • Author(s)
      Uwano Fumito、Kitajima Eiki、Takadama Keiki
    • Journal Title

      Transactions of the Japanese Society for Artificial Intelligence

      Volume: 36 Issue: 6 Pages: B-KB2_1-12

    • DOI

      10.1527/tjsai.36-6_B-KB2

    • NAID

      130008110422

    • ISSN
      1346-0714, 1346-8030
    • Year and Date
      2021-11-01
    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Presentation] LSTM-based Abstraction of Hetero Observation and Transition in Non-Communicative Multi-Agent Reinforcement Learning2022

    • Author(s)
      Fumito Uwano
    • Organizer
      14th International Conference on Agents and Artificial Intelligence (ICAART 2022)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Policy-oriented Goal Selection in Multi-Agent Reinforcement Learning for Dynamic Environments without Communication2022

    • Author(s)
      Fumito Uwano
    • Organizer
      27th International Symposium on Artificial Life and Robotics (AROB 2022)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] A Cooperative Learning Method for Multi-Agent System with Different Input Resolutions2021

    • Author(s)
      Fumito Uwano
    • Organizer
      4th International Symposium on Agents, Multi-Agent Systems and Robotics (ISAMSR 2021)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] マルチエージェントシステムにおける協調行動の抽象度と深層強化学習器の関係性の考察2021

    • Author(s)
      上野 史
    • Organizer
      第48回知能システムシンポジウム
    • Related Report
      2020 Research-status Report
  • [Presentation] 動的環境におけるマルチエージェント強化学習―不完全な情報から集団を動かす仕組み―2021

    • Author(s)
      上野 史
    • Organizer
      第6回岡山大学AI研究会
    • Related Report
      2020 Research-status Report
    • Invited

URL: 

Published: 2020-09-29   Modified: 2023-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi