Improvement for action strategy learning in classification task using classification probalilities

Chyon Hae Kim, Shota Yamazaki, Hiroshi Tsujino, Shigeki Sugano

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

    2 Citations (Scopus)

    Abstract

    In this paper, we address the autonomous evidence accumulation when a system classifies an object into one of predetermined categories. We propose a reinforcement learning system that effectively selects actions to speed up the classification process. The proposed system accelerates its learning using classification probabilities calculated by a classification system. We conducted three binary classification experiments to evaluate the learning speed and correctness of the proposed system. In the first experiment, we examined a random action selection strategy that does not learn its selection parameters while accumulating evidence. In the second experiment, we examined Paletta's reinforcement learning system that observes the state of the object and learns action selection strategy. In the third experiment, we examined the proposed system that observes both the object state and the classification probability. The proposed system showed the fastest learning.

    Original languageEnglish
    Title of host publication6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012
    Pages352-358
    Number of pages7
    DOIs
    Publication statusPublished - 2012
    Event2012 Joint 6th International Conference on Soft Computing and Intelligent Systems, SCIS 2012 and 13th International Symposium on Advanced Intelligence Systems, ISIS 2012 - Kobe
    Duration: 2012 Nov 202012 Nov 24

    Other

    Other2012 Joint 6th International Conference on Soft Computing and Intelligent Systems, SCIS 2012 and 13th International Symposium on Advanced Intelligence Systems, ISIS 2012
    CityKobe
    Period12/11/2012/11/24

    Fingerprint

    Reinforcement learning
    Learning systems
    Experiments

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Software

    Cite this

    Kim, C. H., Yamazaki, S., Tsujino, H., & Sugano, S. (2012). Improvement for action strategy learning in classification task using classification probalilities. In 6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012 (pp. 352-358). [6505012] https://doi.org/10.1109/SCIS-ISIS.2012.6505012

    Improvement for action strategy learning in classification task using classification probalilities. / Kim, Chyon Hae; Yamazaki, Shota; Tsujino, Hiroshi; Sugano, Shigeki.

    6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012. 2012. p. 352-358 6505012.

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

    Kim, CH, Yamazaki, S, Tsujino, H & Sugano, S 2012, Improvement for action strategy learning in classification task using classification probalilities. in 6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012., 6505012, pp. 352-358, 2012 Joint 6th International Conference on Soft Computing and Intelligent Systems, SCIS 2012 and 13th International Symposium on Advanced Intelligence Systems, ISIS 2012, Kobe, 12/11/20. https://doi.org/10.1109/SCIS-ISIS.2012.6505012
    Kim CH, Yamazaki S, Tsujino H, Sugano S. Improvement for action strategy learning in classification task using classification probalilities. In 6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012. 2012. p. 352-358. 6505012 https://doi.org/10.1109/SCIS-ISIS.2012.6505012
    Kim, Chyon Hae ; Yamazaki, Shota ; Tsujino, Hiroshi ; Sugano, Shigeki. / Improvement for action strategy learning in classification task using classification probalilities. 6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012. 2012. pp. 352-358
    @inproceedings{87a656098e584101807f44f1bd156ccc,
    title = "Improvement for action strategy learning in classification task using classification probalilities",
    abstract = "In this paper, we address the autonomous evidence accumulation when a system classifies an object into one of predetermined categories. We propose a reinforcement learning system that effectively selects actions to speed up the classification process. The proposed system accelerates its learning using classification probabilities calculated by a classification system. We conducted three binary classification experiments to evaluate the learning speed and correctness of the proposed system. In the first experiment, we examined a random action selection strategy that does not learn its selection parameters while accumulating evidence. In the second experiment, we examined Paletta's reinforcement learning system that observes the state of the object and learns action selection strategy. In the third experiment, we examined the proposed system that observes both the object state and the classification probability. The proposed system showed the fastest learning.",
    author = "Kim, {Chyon Hae} and Shota Yamazaki and Hiroshi Tsujino and Shigeki Sugano",
    year = "2012",
    doi = "10.1109/SCIS-ISIS.2012.6505012",
    language = "English",
    isbn = "9781467327428",
    pages = "352--358",
    booktitle = "6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012",

    }

    TY - GEN

    T1 - Improvement for action strategy learning in classification task using classification probalilities

    AU - Kim, Chyon Hae

    AU - Yamazaki, Shota

    AU - Tsujino, Hiroshi

    AU - Sugano, Shigeki

    PY - 2012

    Y1 - 2012

    N2 - In this paper, we address the autonomous evidence accumulation when a system classifies an object into one of predetermined categories. We propose a reinforcement learning system that effectively selects actions to speed up the classification process. The proposed system accelerates its learning using classification probabilities calculated by a classification system. We conducted three binary classification experiments to evaluate the learning speed and correctness of the proposed system. In the first experiment, we examined a random action selection strategy that does not learn its selection parameters while accumulating evidence. In the second experiment, we examined Paletta's reinforcement learning system that observes the state of the object and learns action selection strategy. In the third experiment, we examined the proposed system that observes both the object state and the classification probability. The proposed system showed the fastest learning.

    AB - In this paper, we address the autonomous evidence accumulation when a system classifies an object into one of predetermined categories. We propose a reinforcement learning system that effectively selects actions to speed up the classification process. The proposed system accelerates its learning using classification probabilities calculated by a classification system. We conducted three binary classification experiments to evaluate the learning speed and correctness of the proposed system. In the first experiment, we examined a random action selection strategy that does not learn its selection parameters while accumulating evidence. In the second experiment, we examined Paletta's reinforcement learning system that observes the state of the object and learns action selection strategy. In the third experiment, we examined the proposed system that observes both the object state and the classification probability. The proposed system showed the fastest learning.

    UR - http://www.scopus.com/inward/record.url?scp=84877814194&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84877814194&partnerID=8YFLogxK

    U2 - 10.1109/SCIS-ISIS.2012.6505012

    DO - 10.1109/SCIS-ISIS.2012.6505012

    M3 - Conference contribution

    AN - SCOPUS:84877814194

    SN - 9781467327428

    SP - 352

    EP - 358

    BT - 6th International Conference on Soft Computing and Intelligent Systems, and 13th International Symposium on Advanced Intelligence Systems, SCIS/ISIS 2012

    ER -