Summary
International Symposium on Nonlinear Theory and its Applications
2005
Session Number:1-3-3
Session:
Number:1-3-3-1
A Detection Method of Environmental Changes Using Recurrence Plots for Reinforcement Learning
Tetsuya Takahashi, Masaharu Adachi,
pp.246-249
Publication Date:2005/10/18
Online ISSN:2188-5079
DOI:10.34385/proc.40.1-3-3-1
PDF download (892.1KB)
Summary:
In this article, we propose a novel method for detecting environmental changes in the reinforcement learning. The proposed method utilizes recurrence plots of state transitions of the system, and quantifies changes of the recurrence plot by a texture analysis. It is shown that the proposed method is effective to detect environmental changes.