Presentation 2008-10-23
Application of Q-Routing with parameter control to dynamic network environments
Shunsuke ARAKI, Masaharu ADACHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, scales of Wide Area Network (WAN) and Backbone Network are growing because of increase of the number of network users. However, it becomes more difficult to manage the traffic control in centralized way as the scale of the network grows. It is difficult to adjust traffics to dynamic network environments (an increase in network loading). Q-Routing is an autonomous distributed routing approach based on the framework of reinforcement learning, and the capacity for adapting to dynamic network environments is known to be high. In this report, we propose a parameter control method for the Q-Routing and show that it improves the capacity of the method for adapting to dynamic network environments by numerical experiments.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Reinforcement learning / Routing / Dynamic environment / Parameter control
Paper # NC2008-36
Date of Issue

Conference Information
Committee NC
Conference Date 2008/10/16(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Application of Q-Routing with parameter control to dynamic network environments
Sub Title (in English)
Keyword(1) Reinforcement learning
Keyword(2) Routing
Keyword(3) Dynamic environment
Keyword(4) Parameter control
1st Author's Name Shunsuke ARAKI
1st Author's Affiliation Department of Electronic Engineering, Graduate School of Engineering, Tokyo Denki University()
2nd Author's Name Masaharu ADACHI
2nd Author's Affiliation Department of Electronic Engineering, Graduate School of Engineering, Tokyo Denki University
Date 2008-10-23
Paper # NC2008-36
Volume (vol) vol.108
Number (no) 264
Page pp.pp.-
#Pages 6
Date of Issue