Presentation 2005-06-23
Improvement of State Generator Based on ART Neural Network for Reinforcement Learning
Taisuke Nakamura, Takeshi Kamio, Kunihiko Mitsubori, Hisato Fujisaka,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The trade-off between exploration and exploitation has often been discussed in studies on reinforcement learning (RL). This is because exploration and exploitation influence the quality of solutions and the learning efficiency respectively. Previously, we have proposed an adaptive state space segmentation method based on ART neural network for RL. This method is useful for not only the state space segmentation but also the balance between exploration and exploitation. However, if the exploration strength is too large, the learning efficiency degreases rapidly. Since the appropriate strength is generally unknown, this problem must be solved. In this report, we propose a new segmentation method based on ART with two learning phases to improve our conventional method in the tolerance of exploration strength.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Reinforcement Learning / ART Neural Network / State Space / Exploration / Exploitation
Paper # NLP2005-20
Date of Issue

Conference Information
Committee NLP
Conference Date 2005/6/16(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Improvement of State Generator Based on ART Neural Network for Reinforcement Learning
Sub Title (in English)
Keyword(1) Reinforcement Learning
Keyword(2) ART Neural Network
Keyword(3) State Space
Keyword(4) Exploration
Keyword(5) Exploitation
1st Author's Name Taisuke Nakamura
1st Author's Affiliation Hiroshima City University()
2nd Author's Name Takeshi Kamio
2nd Author's Affiliation Hiroshima City University
3rd Author's Name Kunihiko Mitsubori
3rd Author's Affiliation Japan Coast Guard Academy
4th Author's Name Hisato Fujisaka
4th Author's Affiliation Hiroshima City University
Date 2005-06-23
Paper # NLP2005-20
Volume (vol) vol.105
Number (no) 125
Page pp.pp.-
#Pages 6
Date of Issue