Presentation 2012-07-30
Determination of the Change Timing of Space Segmentation Using the Entropy for Reinforcement Learning
Yuki KOMORI, Akira NOTSU, Katsuhiro HONDA, Hidetomo ICHIHASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We tested a single pendulum simulation and observed the influence of the several situation space segmentation patterns in reinforcement learning processes in order to propose new determination of the change timing of space segmentation. Its segmentation is performed by Segmentation and Integration method or Contraction Method. Additionally, the entropy, which was defined on action values' distributions was used to get the timing of the changing space segmentation. Simulation results were shown to demonstrate the influence and adaptability of the proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Reinforcement learning / Space segmentation / Entropy
Paper # NC2012-15
Date of Issue

Conference Information
Committee NC
Conference Date 2012/7/23(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Determination of the Change Timing of Space Segmentation Using the Entropy for Reinforcement Learning
Sub Title (in English)
Keyword(1) Reinforcement learning
Keyword(2) Space segmentation
Keyword(3) Entropy
1st Author's Name Yuki KOMORI
1st Author's Affiliation Graduate School of Engineering, Osaka Prefecture University()
2nd Author's Name Akira NOTSU
2nd Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
3rd Author's Name Katsuhiro HONDA
3rd Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
4th Author's Name Hidetomo ICHIHASHI
4th Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
Date 2012-07-30
Paper # NC2012-15
Volume (vol) vol.112
Number (no) 168
Page pp.pp.-
#Pages 4
Date of Issue