Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty

Ryoichi Nakajo, Maasa Takahashi, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

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

    2 Citations (Scopus)

    Abstract

    In this paper, we propose a model that can explain the mechanism of self and non-self discrimination. Infants gradually develop their abilities for self–other cognition through interaction with the environment. Predictive learning has been widely used to explain the mechanism of infants’ development. We hypothesized that infants’ cognitive abilities are developed through predictive learning and the uncertainty estimation of their sensory-motor inputs. We chose a stochastic continuous time recurrent neural network, which is a dynamical neural network model, to predict uncertainties as variances. From the perspective of cognitive developmental robotics, a predictive learning experiment with a robot was performed. The results indicate that training made the robot predict the regions related to its body more easily. We confirmed that self and non-self cognitive abilities might be acquired through predictive learning with uncertainty estimation.

    Original languageEnglish
    Title of host publicationNeural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings
    PublisherSpringer Verlag
    Pages228-235
    Number of pages8
    Volume9950 LNCS
    ISBN (Print)9783319466804
    DOIs
    Publication statusPublished - 2016
    Event23rd International Conference on Neural Information Processing, ICONIP 2016 - Kyoto, Japan
    Duration: 2016 Oct 162016 Oct 21

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume9950 LNCS
    ISSN (Print)03029743
    ISSN (Electronic)16113349

    Other

    Other23rd International Conference on Neural Information Processing, ICONIP 2016
    CountryJapan
    CityKyoto
    Period16/10/1616/10/21

    Fingerprint

    Discrimination
    Uncertainty Estimation
    Uncertainty
    Robots
    Recurrent neural networks
    Robot
    Predict
    Recurrent Neural Networks
    Robotics
    Cognition
    Neural Network Model
    Neural networks
    Continuous Time
    Choose
    Learning
    Experiments
    Interaction
    Experiment
    Model

    Keywords

    • Cognitive developmental robotics
    • Recurrent neural network
    • Self/non-self cognition

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • Computer Science(all)

    Cite this

    Nakajo, R., Takahashi, M., Murata, S., Arie, H., & Ogata, T. (2016). Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty. In Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings (Vol. 9950 LNCS, pp. 228-235). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9950 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-46681-1_28

    Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty. / Nakajo, Ryoichi; Takahashi, Maasa; Murata, Shingo; Arie, Hiroaki; Ogata, Tetsuya.

    Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Vol. 9950 LNCS Springer Verlag, 2016. p. 228-235 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9950 LNCS).

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

    Nakajo, R, Takahashi, M, Murata, S, Arie, H & Ogata, T 2016, Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty. in Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. vol. 9950 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9950 LNCS, Springer Verlag, pp. 228-235, 23rd International Conference on Neural Information Processing, ICONIP 2016, Kyoto, Japan, 16/10/16. https://doi.org/10.1007/978-3-319-46681-1_28
    Nakajo R, Takahashi M, Murata S, Arie H, Ogata T. Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty. In Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Vol. 9950 LNCS. Springer Verlag. 2016. p. 228-235. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-46681-1_28
    Nakajo, Ryoichi ; Takahashi, Maasa ; Murata, Shingo ; Arie, Hiroaki ; Ogata, Tetsuya. / Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty. Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings. Vol. 9950 LNCS Springer Verlag, 2016. pp. 228-235 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
    @inproceedings{6da38907e9f349308f628a748e2310c5,
    title = "Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty",
    abstract = "In this paper, we propose a model that can explain the mechanism of self and non-self discrimination. Infants gradually develop their abilities for self–other cognition through interaction with the environment. Predictive learning has been widely used to explain the mechanism of infants’ development. We hypothesized that infants’ cognitive abilities are developed through predictive learning and the uncertainty estimation of their sensory-motor inputs. We chose a stochastic continuous time recurrent neural network, which is a dynamical neural network model, to predict uncertainties as variances. From the perspective of cognitive developmental robotics, a predictive learning experiment with a robot was performed. The results indicate that training made the robot predict the regions related to its body more easily. We confirmed that self and non-self cognitive abilities might be acquired through predictive learning with uncertainty estimation.",
    keywords = "Cognitive developmental robotics, Recurrent neural network, Self/non-self cognition",
    author = "Ryoichi Nakajo and Maasa Takahashi and Shingo Murata and Hiroaki Arie and Tetsuya Ogata",
    year = "2016",
    doi = "10.1007/978-3-319-46681-1_28",
    language = "English",
    isbn = "9783319466804",
    volume = "9950 LNCS",
    series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
    publisher = "Springer Verlag",
    pages = "228--235",
    booktitle = "Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings",
    address = "Germany",

    }

    TY - GEN

    T1 - Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty

    AU - Nakajo, Ryoichi

    AU - Takahashi, Maasa

    AU - Murata, Shingo

    AU - Arie, Hiroaki

    AU - Ogata, Tetsuya

    PY - 2016

    Y1 - 2016

    N2 - In this paper, we propose a model that can explain the mechanism of self and non-self discrimination. Infants gradually develop their abilities for self–other cognition through interaction with the environment. Predictive learning has been widely used to explain the mechanism of infants’ development. We hypothesized that infants’ cognitive abilities are developed through predictive learning and the uncertainty estimation of their sensory-motor inputs. We chose a stochastic continuous time recurrent neural network, which is a dynamical neural network model, to predict uncertainties as variances. From the perspective of cognitive developmental robotics, a predictive learning experiment with a robot was performed. The results indicate that training made the robot predict the regions related to its body more easily. We confirmed that self and non-self cognitive abilities might be acquired through predictive learning with uncertainty estimation.

    AB - In this paper, we propose a model that can explain the mechanism of self and non-self discrimination. Infants gradually develop their abilities for self–other cognition through interaction with the environment. Predictive learning has been widely used to explain the mechanism of infants’ development. We hypothesized that infants’ cognitive abilities are developed through predictive learning and the uncertainty estimation of their sensory-motor inputs. We chose a stochastic continuous time recurrent neural network, which is a dynamical neural network model, to predict uncertainties as variances. From the perspective of cognitive developmental robotics, a predictive learning experiment with a robot was performed. The results indicate that training made the robot predict the regions related to its body more easily. We confirmed that self and non-self cognitive abilities might be acquired through predictive learning with uncertainty estimation.

    KW - Cognitive developmental robotics

    KW - Recurrent neural network

    KW - Self/non-self cognition

    UR - http://www.scopus.com/inward/record.url?scp=84992723857&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84992723857&partnerID=8YFLogxK

    U2 - 10.1007/978-3-319-46681-1_28

    DO - 10.1007/978-3-319-46681-1_28

    M3 - Conference contribution

    SN - 9783319466804

    VL - 9950 LNCS

    T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

    SP - 228

    EP - 235

    BT - Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings

    PB - Springer Verlag

    ER -