since 07 November 2024 :

View(s) :

13 (0 ULiège)

Download(s) :

5 (0 ULiège)

References
Facsimile [PDF, 2.9M]

Measuring Information Transfer in Online Adaptation Process of Recurrent Neural Networks

Kohei Nakajima

p. 33-41

    Abstract

    In this paper, we propose a simple model that focuses on the adaptation process of an agent to an unknown system in an online manner. The agent is equipped with a recurrent neural network, and by controlling the dynamics of the interacting system, it should predict its state in one-step prediction. To quantitatively characterize the interaction modality between the agent and the interacting system, we used transfer entropy. As a result, by varying the nonlinear parameter of the interacting system and the coupling strength, we numerically show that the adaptation dynamics can be distinguished between an agent-driven and a non-agent-driven dynamics.

    Index

    Keywords

    recurrent neural network, online learning, transfer entropy

    Text

    Download Facsimile [PDF, 2.9M]

    References

    Bibliographical reference

    Kohei Nakajima, « Measuring Information Transfer in Online Adaptation Process of Recurrent Neural Networks », CASYS, 24 | 2010, 33-41.

    Electronic reference

    Kohei Nakajima, « Measuring Information Transfer in Online Adaptation Process of Recurrent Neural Networks », CASYS [Online], 24 | 2010, Online since 06 September 2024, connection on 02 June 2025. URL : http://popups.uliege.be/3041-539x/index.php?id=3023

    Author

    Kohei Nakajima

    Department of General Systems Sciences, The Graduate School of Arts and Sciences, The University of Tokyo, 3-8-1 Komaba, Tokyo 153-8902, Japan

    By this author

    Copyright

    CC BY-SA 4.0 Deed