Learning to generate proactive and reactive behavior using a dynamic neural network model with time-varying variance prediction mechanism

Shingo Murata, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano, Jun Tani

    Research output: Contribution to journalArticle

    5 Citations (Scopus)

    Abstract

    This paper discusses a possible neurodynamic mechanism that enables self-organization of two basic behavioral modes, namely a proactive mode and a reactive mode, and of autonomous switching between these modes depending on the situation. In the proactive mode, actions are generated based on an internal prediction, whereas in the reactive mode actions are generated in response to sensory inputs in unpredictable situations. In order to investigate how these two behavioral modes can be self-organized and how autonomous switching between the two modes can be achieved, we conducted neurorobotics experiments by using our recently developed dynamic neural network model that has a capability to learn to predict time-varying variance of the observable variables. In a set of robot experiments under various conditions, the robot was required to imitate others movements consisting of alternating predictable and unpredictable patterns. The experimental results showed that the robot controlled by the neural network model was able to proactively imitate predictable patterns and reactively follow unpredictable patterns by autonomously switching its behavioral modes. Our analysis revealed that variance prediction mechanism can lead to self-organization of these abilities with sufficient robustness and generalization capabilities.

    Original languageEnglish
    Pages (from-to)1189-1203
    Number of pages15
    JournalAdvanced Robotics
    Volume28
    Issue number17
    DOIs
    Publication statusPublished - 2014 Sep 2

    Fingerprint

    Robots
    Neural networks
    Experiments

    Keywords

    • humanoid robot
    • imitation
    • proactive behavior
    • reactive behavior
    • recurrent neural network

    ASJC Scopus subject areas

    • Control and Systems Engineering
    • Human-Computer Interaction
    • Computer Science Applications
    • Hardware and Architecture
    • Software

    Cite this

    Learning to generate proactive and reactive behavior using a dynamic neural network model with time-varying variance prediction mechanism. / Murata, Shingo; Arie, Hiroaki; Ogata, Tetsuya; Sugano, Shigeki; Tani, Jun.

    In: Advanced Robotics, Vol. 28, No. 17, 02.09.2014, p. 1189-1203.

    Research output: Contribution to journalArticle

    @article{82e02d164a6f41d3884cbe979764d22e,
    title = "Learning to generate proactive and reactive behavior using a dynamic neural network model with time-varying variance prediction mechanism",
    abstract = "This paper discusses a possible neurodynamic mechanism that enables self-organization of two basic behavioral modes, namely a proactive mode and a reactive mode, and of autonomous switching between these modes depending on the situation. In the proactive mode, actions are generated based on an internal prediction, whereas in the reactive mode actions are generated in response to sensory inputs in unpredictable situations. In order to investigate how these two behavioral modes can be self-organized and how autonomous switching between the two modes can be achieved, we conducted neurorobotics experiments by using our recently developed dynamic neural network model that has a capability to learn to predict time-varying variance of the observable variables. In a set of robot experiments under various conditions, the robot was required to imitate others movements consisting of alternating predictable and unpredictable patterns. The experimental results showed that the robot controlled by the neural network model was able to proactively imitate predictable patterns and reactively follow unpredictable patterns by autonomously switching its behavioral modes. Our analysis revealed that variance prediction mechanism can lead to self-organization of these abilities with sufficient robustness and generalization capabilities.",
    keywords = "humanoid robot, imitation, proactive behavior, reactive behavior, recurrent neural network",
    author = "Shingo Murata and Hiroaki Arie and Tetsuya Ogata and Shigeki Sugano and Jun Tani",
    year = "2014",
    month = "9",
    day = "2",
    doi = "10.1080/01691864.2014.916628",
    language = "English",
    volume = "28",
    pages = "1189--1203",
    journal = "Advanced Robotics",
    issn = "0169-1864",
    publisher = "Taylor and Francis Ltd.",
    number = "17",

    }

    TY - JOUR

    T1 - Learning to generate proactive and reactive behavior using a dynamic neural network model with time-varying variance prediction mechanism

    AU - Murata, Shingo

    AU - Arie, Hiroaki

    AU - Ogata, Tetsuya

    AU - Sugano, Shigeki

    AU - Tani, Jun

    PY - 2014/9/2

    Y1 - 2014/9/2

    N2 - This paper discusses a possible neurodynamic mechanism that enables self-organization of two basic behavioral modes, namely a proactive mode and a reactive mode, and of autonomous switching between these modes depending on the situation. In the proactive mode, actions are generated based on an internal prediction, whereas in the reactive mode actions are generated in response to sensory inputs in unpredictable situations. In order to investigate how these two behavioral modes can be self-organized and how autonomous switching between the two modes can be achieved, we conducted neurorobotics experiments by using our recently developed dynamic neural network model that has a capability to learn to predict time-varying variance of the observable variables. In a set of robot experiments under various conditions, the robot was required to imitate others movements consisting of alternating predictable and unpredictable patterns. The experimental results showed that the robot controlled by the neural network model was able to proactively imitate predictable patterns and reactively follow unpredictable patterns by autonomously switching its behavioral modes. Our analysis revealed that variance prediction mechanism can lead to self-organization of these abilities with sufficient robustness and generalization capabilities.

    AB - This paper discusses a possible neurodynamic mechanism that enables self-organization of two basic behavioral modes, namely a proactive mode and a reactive mode, and of autonomous switching between these modes depending on the situation. In the proactive mode, actions are generated based on an internal prediction, whereas in the reactive mode actions are generated in response to sensory inputs in unpredictable situations. In order to investigate how these two behavioral modes can be self-organized and how autonomous switching between the two modes can be achieved, we conducted neurorobotics experiments by using our recently developed dynamic neural network model that has a capability to learn to predict time-varying variance of the observable variables. In a set of robot experiments under various conditions, the robot was required to imitate others movements consisting of alternating predictable and unpredictable patterns. The experimental results showed that the robot controlled by the neural network model was able to proactively imitate predictable patterns and reactively follow unpredictable patterns by autonomously switching its behavioral modes. Our analysis revealed that variance prediction mechanism can lead to self-organization of these abilities with sufficient robustness and generalization capabilities.

    KW - humanoid robot

    KW - imitation

    KW - proactive behavior

    KW - reactive behavior

    KW - recurrent neural network

    UR - http://www.scopus.com/inward/record.url?scp=84904978388&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84904978388&partnerID=8YFLogxK

    U2 - 10.1080/01691864.2014.916628

    DO - 10.1080/01691864.2014.916628

    M3 - Article

    AN - SCOPUS:84904978388

    VL - 28

    SP - 1189

    EP - 1203

    JO - Advanced Robotics

    JF - Advanced Robotics

    SN - 0169-1864

    IS - 17

    ER -