An asymmetric multi-agent learning model and its simulation analysis

Haiyan Qiao, Jerzy W Rozenblit, Ferenc Szidarovszky

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Multi-agent decision problems in unknown en- vironments are common where the agents are usually empowered with di®erent decision pow- ers and involved in some sort of the prisoner's dilemma problem. A general solution to this kind of complex decision problem is that the agents cooperate to play a joint action. Asym- metric Nash bargaining solution is an attractive approach to such cooperative games with players of di®erent powers. In this paper, a new multi- agent learning algorithm based on the asymmet- ric Nash bargaining solution is presented. Sim- ulation is performed on a testbed of stochastic games. The experimental results demonstrate that the algorithm is fast and converges to a Pareto-optimal solution. Compared with the learning algorithms based on non-cooperative equilibrium, this approach is faster and avoids the disturbing problem of equilibrium selection.

Original languageEnglish (US)
Title of host publicationInternational Mediterranean Modelling Multiconference, I3M
Pages231-237
Number of pages7
StatePublished - 2006
EventInternational Mediterranean Modelling Multiconference, I3M 2006 - Barcelona, Spain
Duration: Oct 4 2006Oct 6 2006

Other

OtherInternational Mediterranean Modelling Multiconference, I3M 2006
CountrySpain
CityBarcelona
Period10/4/0610/6/06

Fingerprint

Nash Bargaining Solution
Multiagent Learning
Simulation Analysis
Decision problem
Learning algorithms
Learning Algorithm
Equilibrium Selection
Prisoners' Dilemma
Stochastic Games
Cooperative Game
Pareto Optimal Solution
Testbeds
General Solution
Testbed
Sort
Converge
Metric
Unknown
Experimental Results
Model

Keywords

  • Asymmetric Nash bargaining solution
  • Multi- agent learning
  • Pareto-optimality
  • Simulation
  • Social dilemma

ASJC Scopus subject areas

  • Modeling and Simulation

Cite this

Qiao, H., Rozenblit, J. W., & Szidarovszky, F. (2006). An asymmetric multi-agent learning model and its simulation analysis. In International Mediterranean Modelling Multiconference, I3M (pp. 231-237)

An asymmetric multi-agent learning model and its simulation analysis. / Qiao, Haiyan; Rozenblit, Jerzy W; Szidarovszky, Ferenc.

International Mediterranean Modelling Multiconference, I3M. 2006. p. 231-237.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Qiao, H, Rozenblit, JW & Szidarovszky, F 2006, An asymmetric multi-agent learning model and its simulation analysis. in International Mediterranean Modelling Multiconference, I3M. pp. 231-237, International Mediterranean Modelling Multiconference, I3M 2006, Barcelona, Spain, 10/4/06.
Qiao H, Rozenblit JW, Szidarovszky F. An asymmetric multi-agent learning model and its simulation analysis. In International Mediterranean Modelling Multiconference, I3M. 2006. p. 231-237
Qiao, Haiyan ; Rozenblit, Jerzy W ; Szidarovszky, Ferenc. / An asymmetric multi-agent learning model and its simulation analysis. International Mediterranean Modelling Multiconference, I3M. 2006. pp. 231-237
@inproceedings{039c1e2c118f4071985088f0105a78c9,
title = "An asymmetric multi-agent learning model and its simulation analysis",
abstract = "Multi-agent decision problems in unknown en- vironments are common where the agents are usually empowered with di{\circledR}erent decision pow- ers and involved in some sort of the prisoner's dilemma problem. A general solution to this kind of complex decision problem is that the agents cooperate to play a joint action. Asym- metric Nash bargaining solution is an attractive approach to such cooperative games with players of di{\circledR}erent powers. In this paper, a new multi- agent learning algorithm based on the asymmet- ric Nash bargaining solution is presented. Sim- ulation is performed on a testbed of stochastic games. The experimental results demonstrate that the algorithm is fast and converges to a Pareto-optimal solution. Compared with the learning algorithms based on non-cooperative equilibrium, this approach is faster and avoids the disturbing problem of equilibrium selection.",
keywords = "Asymmetric Nash bargaining solution, Multi- agent learning, Pareto-optimality, Simulation, Social dilemma",
author = "Haiyan Qiao and Rozenblit, {Jerzy W} and Ferenc Szidarovszky",
year = "2006",
language = "English (US)",
isbn = "8469007262",
pages = "231--237",
booktitle = "International Mediterranean Modelling Multiconference, I3M",

}

TY - GEN

T1 - An asymmetric multi-agent learning model and its simulation analysis

AU - Qiao, Haiyan

AU - Rozenblit, Jerzy W

AU - Szidarovszky, Ferenc

PY - 2006

Y1 - 2006

N2 - Multi-agent decision problems in unknown en- vironments are common where the agents are usually empowered with di®erent decision pow- ers and involved in some sort of the prisoner's dilemma problem. A general solution to this kind of complex decision problem is that the agents cooperate to play a joint action. Asym- metric Nash bargaining solution is an attractive approach to such cooperative games with players of di®erent powers. In this paper, a new multi- agent learning algorithm based on the asymmet- ric Nash bargaining solution is presented. Sim- ulation is performed on a testbed of stochastic games. The experimental results demonstrate that the algorithm is fast and converges to a Pareto-optimal solution. Compared with the learning algorithms based on non-cooperative equilibrium, this approach is faster and avoids the disturbing problem of equilibrium selection.

AB - Multi-agent decision problems in unknown en- vironments are common where the agents are usually empowered with di®erent decision pow- ers and involved in some sort of the prisoner's dilemma problem. A general solution to this kind of complex decision problem is that the agents cooperate to play a joint action. Asym- metric Nash bargaining solution is an attractive approach to such cooperative games with players of di®erent powers. In this paper, a new multi- agent learning algorithm based on the asymmet- ric Nash bargaining solution is presented. Sim- ulation is performed on a testbed of stochastic games. The experimental results demonstrate that the algorithm is fast and converges to a Pareto-optimal solution. Compared with the learning algorithms based on non-cooperative equilibrium, this approach is faster and avoids the disturbing problem of equilibrium selection.

KW - Asymmetric Nash bargaining solution

KW - Multi- agent learning

KW - Pareto-optimality

KW - Simulation

KW - Social dilemma

UR - http://www.scopus.com/inward/record.url?scp=84870208555&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870208555&partnerID=8YFLogxK

M3 - Conference contribution

SN - 8469007262

SN - 9788469007266

SP - 231

EP - 237

BT - International Mediterranean Modelling Multiconference, I3M

ER -