NAGAO Tomoharu

Affiliation

Faculty of Environment and Information Sciences, Division of Social Environment and Information

Job Title

Professor

Date of Birth

1959

Research Fields, Keywords

Artificial Intelligence,Evolutionary Computation,Image Processing

Mail Address

E-mail address



写真a

The Best Research Acheivement as Researcher Life 【 display / non-display

Graduate School 【 display / non-display

  •  
    -
    1985.05

    Tokyo Institute of Technology  Graduate School, Division of Integrated Science and Engineering  Department of Information Processing  Doctor Course  Unfinished course

External Career 【 display / non-display

  • 1995.02
    -
    2000.06

    Tokyo Institute of Technology   Imaging Science and Engineering Laboratory   Associate Professor (as old post name)  

  • 1985.06
    -
    1995.01

    Imaging Science and Engineering Laboratory, Tokyo Institute of Technology   Research Assistant  

Field of expertise (Grants-in-aid for Scientific Research classification) 【 display / non-display

  • Intelligent informatics

  • Perceptual information processing

  • Kansei informatics

  • Intelligent robotics

 

Research Career 【 display / non-display

  • next-generation machine learning

    Project Year:  -   

  • advanced neural networks

    Project Year:  -   

  • KANSEI information processing and its applications

    Project Year:  -   

  • generalized evolutionary computation

    Project Year:  -   

Books 【 display / non-display

  • Genetic Algorithm

    Takeshi Agui (Part: Joint Work )

    Syokodo  1993.09

  • Image Processing and Recognition

    Takeshi Agui (Part: Joint Work )

    Syokodo  1992.04

Papers 【 display / non-display

  • Automatic Generation of Sentences Explaining Image Classification Processes Constructed by Decision Tree and Decision Network

    Miho Sakitsu, Daiki Tsuchiya, Masanori Suganuma, Tomoharu Nagao

    Transactions on mathematical modeling and its applications ( Information Processing Society of Japan )  9 ( 1 )   43 - 52   2016.02  [Refereed]

    Joint Work

  • A Self-Organising Model for Anomaly Detection and Its Application to Video Surveillance

    Masanori Suganuma, Tomoharu Nagao

    Transactions on mathematical modeling and its applications ( Information Processing Society of Japan )  9 ( 1 )   23 - 32   2016.02  [Refereed]

    Joint Work

    CiNii

  • Evolutionary Image Processing Application to Painting

    Midori Saito, Tomoharu Nagao

    Transactions on mathematical modeling and its applications ( Information Processing Society of Japan )  8 ( 3 )   26 - 35   2015.11

    Joint Work

    CiNii

  • Image Classification Based on Hierarchical Feature Dimension Reduction Using Cartesian Genetic Programming

    Kota Saito, Tomoharu Nagao

    Transaction of the Japanese Society for Evolutionary Computation ( The Japanese Society for Evolutionary Computation )  6 ( 2 )   55 - 66   2015  [Refereed]

    Joint Work

    DOI CiNii

  • Controlling an Autonomous Agent for Exploring Unknown Environments using Switching Prelearned Modules

    Hata Takahito, Suganuma Masanori, Nagao Tomoharu

    IEEJ Transactions on Electronics, Information and Systems ( The Institute of Electrical Engineers of Japan )  138 ( 2 )   157 - 164   2018

    Joint Work

     View Summary

    <p>In this paper, we try to acquire various behavior patterns of autonomous exploration agent using several learning environments. In case of previous learning methods using a single behavior rule set, it is hard to acquire the behavior that covers all learning environments. In our method, we divide learning environments into some primitive environments whose properties differ each other, and then generate modules that are specialized for each primitive environment. To optimize behavior rules of agents, we adopt Graph Structured Program Evolution (GRAPE) which can automatically generates graph structured programs. In unknown environments, each module is switched by a program named "switcher". The switcher selects the module that acts better in a neighboring environment. Through several experiments, our method achieved higher exploration rate in unknown environments compared to simple GRAPE, random search, and the method that switches modules randomly.</p>

    DOI CiNii

display all >>

Preferred joint research theme 【 display / non-display

  • artificial intelligence and its applications

  • intelligent image processing and understanding

  • KANSEI information processing

  • prediction of sequential signals

Past of Collaboration and Commissioned Research 【 display / non-display

  • image processing

    Cooperative Research within Japan  

    Project Year: 2007.10  -   

  • Artificial intelligence in the maritime field

    Cooperative Research within Japan  

    Project Year: 2016.07  -   

  • Optimization of robot control

    Cooperative Research within Japan  

    Project Year: 2016.04  -   

  • Kansei information processing

    Others  

    Project Year: 2013.04  -