Presentation 2003/1/28
Multi-agent Reinforcement Learning on Collective Decision-Making : Designing E-Democracy
Hiroshi YAMAKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, E-Democracy became realistic. In E-Democracy, a flexible system design is possible compared with the conventional democracy. Thereby, the better system which reflects public opinion appropriately may be able to be designed. However, in the present state, the system design technique of collective decision-making is not established. Multi-agent reinforcement learning is one powerful approach carried out to such a design subject. In this report, first, I explain a possibility by new design of E-Democracy system. Second, I present the simple simulation of MA-RL. Third, I compare the subject of collective decision-making and the subject of MA-RL. By the above-mentioned argument, the potentiality of MA-RL as system design technology of collective decision-making was explored.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Direct Democracy / Reinforcement Learning / Collective Decision-Making / Multi-agent System
Paper # NC2002-124
Date of Issue

Conference Information
Committee NC
Conference Date 2003/1/28(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) Multi-agent Reinforcement Learning on Collective Decision-Making : Designing E-Democracy
Sub Title (in English)
Keyword(1) Direct Democracy
Keyword(2) Reinforcement Learning
Keyword(3) Collective Decision-Making
Keyword(4) Multi-agent System
1st Author's Name Hiroshi YAMAKAWA
1st Author's Affiliation Grid Computing & Bioinfomatics Lab. FUJITSU LABORATORIES LTD.()
Date 2003/1/28
Paper # NC2002-124
Volume (vol) vol.102
Number (no) 628
Page pp.pp.-
#Pages 6
Date of Issue