Examination of skill-based learning by inverse reinforcement learning using evolutionary process

Hiroaki Tsunekawa, Takuo Suzuki, Tomoki Hamagami

    Abstract

    In this paper, we propose skill-based learning by inverse reinforcement learning using evolutionary process. Reinforcement learning requires a large amount of time and learning convergence does not depend on the learning targets. In addition, if the learning targets are not known clearly, the appropriate reward cannot be defined and this makes learning difficult. Sub-goal method and inverse reinforcement learning are effective for each problem. They can deal with the problem that it requires a large amount of time and finding appropriate reward is difficult. However, in case that there is interference between behavior rules, the learning is not achieved efficiently by the sub-goal method. Therefore, in this study, the process of learning each behavior rules simultaneously is made with evolutionary process and reward functions for the half way are obtained by inverse reinforcement learning of the process. The target behavior is achieved by using the reward functions. This proposed method is called skill-based learning. Finally, effectiveness of skill-based learning is confirmed by experiment of driving task.

    Original languageEnglish
    Title of host publicationProceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    ISBN (Print)9781479964802
    DOIs
    StatePublished - 2015 Sep 28
    Event9th IEEE International Conference on Intelligent Systems and Control, ISCO 2015 - Coimbatore, India

    Other

    Other9th IEEE International Conference on Intelligent Systems and Control, ISCO 2015
    CountryIndia
    CityCoimbatore
    Period15/1/915/1/10

    Fingerprint

    Reinforcement learning
    Experiments

    Keywords

    • Evolutionary Process
    • Inverse Reinforcement Learning
    • Reinforcement Learning
    • Skill-based Learning

    ASJC Scopus subject areas

    • Computer Science Applications
    • Computer Networks and Communications
    • Artificial Intelligence
    • Electrical and Electronic Engineering
    • Control and Systems Engineering

    Cite this

    Tsunekawa, H., Suzuki, T., & Hamagami, T. (2015). Examination of skill-based learning by inverse reinforcement learning using evolutionary process. In Proceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015. [7282296] Institute of Electrical and Electronics Engineers Inc.. DOI: 10.1109/ISCO.2015.7282296

    Examination of skill-based learning by inverse reinforcement learning using evolutionary process. / Tsunekawa, Hiroaki; Suzuki, Takuo; Hamagami, Tomoki.

    Proceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7282296.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Tsunekawa, H, Suzuki, T & Hamagami, T 2015, Examination of skill-based learning by inverse reinforcement learning using evolutionary process. in Proceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015., 7282296, Institute of Electrical and Electronics Engineers Inc., 9th IEEE International Conference on Intelligent Systems and Control, ISCO 2015, Coimbatore, India, 9-10 January. DOI: 10.1109/ISCO.2015.7282296
    Tsunekawa H, Suzuki T, Hamagami T. Examination of skill-based learning by inverse reinforcement learning using evolutionary process. In Proceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015. Institute of Electrical and Electronics Engineers Inc.2015. 7282296. Available from, DOI: 10.1109/ISCO.2015.7282296

    Tsunekawa, Hiroaki; Suzuki, Takuo; Hamagami, Tomoki / Examination of skill-based learning by inverse reinforcement learning using evolutionary process.

    Proceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7282296.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    @inbook{1a0e7c0f217b458c8cbc5a888aa3bbfe,
    title = "Examination of skill-based learning by inverse reinforcement learning using evolutionary process",
    keywords = "Evolutionary Process, Inverse Reinforcement Learning, Reinforcement Learning, Skill-based Learning",
    author = "Hiroaki Tsunekawa and Takuo Suzuki and Tomoki Hamagami",
    year = "2015",
    month = "9",
    doi = "10.1109/ISCO.2015.7282296",
    isbn = "9781479964802",
    booktitle = "Proceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - CHAP

    T1 - Examination of skill-based learning by inverse reinforcement learning using evolutionary process

    AU - Tsunekawa,Hiroaki

    AU - Suzuki,Takuo

    AU - Hamagami,Tomoki

    PY - 2015/9/28

    Y1 - 2015/9/28

    N2 - In this paper, we propose skill-based learning by inverse reinforcement learning using evolutionary process. Reinforcement learning requires a large amount of time and learning convergence does not depend on the learning targets. In addition, if the learning targets are not known clearly, the appropriate reward cannot be defined and this makes learning difficult. Sub-goal method and inverse reinforcement learning are effective for each problem. They can deal with the problem that it requires a large amount of time and finding appropriate reward is difficult. However, in case that there is interference between behavior rules, the learning is not achieved efficiently by the sub-goal method. Therefore, in this study, the process of learning each behavior rules simultaneously is made with evolutionary process and reward functions for the half way are obtained by inverse reinforcement learning of the process. The target behavior is achieved by using the reward functions. This proposed method is called skill-based learning. Finally, effectiveness of skill-based learning is confirmed by experiment of driving task.

    AB - In this paper, we propose skill-based learning by inverse reinforcement learning using evolutionary process. Reinforcement learning requires a large amount of time and learning convergence does not depend on the learning targets. In addition, if the learning targets are not known clearly, the appropriate reward cannot be defined and this makes learning difficult. Sub-goal method and inverse reinforcement learning are effective for each problem. They can deal with the problem that it requires a large amount of time and finding appropriate reward is difficult. However, in case that there is interference between behavior rules, the learning is not achieved efficiently by the sub-goal method. Therefore, in this study, the process of learning each behavior rules simultaneously is made with evolutionary process and reward functions for the half way are obtained by inverse reinforcement learning of the process. The target behavior is achieved by using the reward functions. This proposed method is called skill-based learning. Finally, effectiveness of skill-based learning is confirmed by experiment of driving task.

    KW - Evolutionary Process

    KW - Inverse Reinforcement Learning

    KW - Reinforcement Learning

    KW - Skill-based Learning

    UR - http://www.scopus.com/inward/record.url?scp=84959162902&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84959162902&partnerID=8YFLogxK

    U2 - 10.1109/ISCO.2015.7282296

    DO - 10.1109/ISCO.2015.7282296

    M3 - Conference contribution

    SN - 9781479964802

    BT - Proceedings of 2015 IEEE 9th International Conference on Intelligent Systems and Control, ISCO 2015

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -