Summary
International Symposium on Nonlinear Theory and its Applications
2010
Session Number:A3L-D
Session:
Number:A3L-D3
Reinforcement Learning using Improved Kohonen Feature Map Probabilistic Associative Memory based on Weights Distribution
Shingo NOGUCHI, Yuko OSANA,
pp.185-188
Publication Date:2010/9/5
Online ISSN:2188-5079
DOI:10.34385/proc.44.A3L-D3
PDF download (208.1KB)
Summary:
In this paper, we propose a reinforcement learning method using Improved Kohonen Feature Map Probabilistic Associative Memory based on Weights Distribution (IKFMPAM-WD). The proposed method is based on the actor-critic method, and the actor is realized by the IKFMPAM-WD. The IKFMPAM-WD is based on the selforganizing feature map, and it can realize successive learning and one-to-many associations. The proposed method makes use of this property in order to realize the learning during the practice of task. We carried out a series of computer experiments, and confirmed the effectiveness of the proposed method in the path-finding problem.