Presentation 2003/3/12
A Model for a Radial-Arms Maze Task using Reinforcement Learning
Susumu Yamazaki, Makoto Ito, Shogo Miyake, Shunji Satoh,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We construct a model of a radial-arms maze task for rats in which a higher function of memory is required. We formulate a radial-arms maze task as a reinforcement learning problem. In this model, agent can successfully learn depnding on past memory by adding feedback to an Actor-Critic method that is one of reinforcement learning algorithms. Simulations show that the model explains a result of the behavioral experiment of a rat well.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) radial-arms maze task / reinforcement learning / Actor-Critic method / feedback
Paper # NC2002-224
Date of Issue

Conference Information
Committee NC
Conference Date 2003/3/12(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) A Model for a Radial-Arms Maze Task using Reinforcement Learning
Sub Title (in English)
Keyword(1) radial-arms maze task
Keyword(2) reinforcement learning
Keyword(3) Actor-Critic method
Keyword(4) feedback
1st Author's Name Susumu Yamazaki
1st Author's Affiliation Department of Applied Physics, Graduate School of Engineering, Tohoku University()
2nd Author's Name Makoto Ito
2nd Author's Affiliation Department of Neurobiology and Behavior, School of Medicine, Gunma University
3rd Author's Name Shogo Miyake
3rd Author's Affiliation Department of Applied Physics, Graduate School of Engineering, Tohoku University
4th Author's Name Shunji Satoh
4th Author's Affiliation Department of Applied Physics, Graduate School of Engineering, Tohoku University
Date 2003/3/12
Paper # NC2002-224
Volume (vol) vol.102
Number (no) 731
Page pp.pp.-
#Pages 6
Date of Issue