Multi-agent learning model with bargaining

Haiyan Qiao, Jerzy W Rozenblit, Ferenc Szidarovszky, Lizhi Yang

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

8 Citations (Scopus)

Abstract

Decision problems with the features of prisoner's dilemma are quite common. A general solution to this kind of social dilemma is that the agents cooperate to play a joint action. The Nash bargaining solution is an attractive approach to such cooperative games. In this paper, a multi-agent learning algorithm based on the Nash bargaining solution is presented. Different experiments are conducted on a testbed of stochastic games. The experimental results demonstrate that the algorithm converges to the policies of the Nash bargaining solution. Compared with the learning algorithms based on a non-cooperative equilibrium, this algorithm is fast and its complexity is linear with respect to the number of agents and number of iterations. In addition, it avoids the disturbing problem of equilibrium selection.

Original languageEnglish (US)
Title of host publicationProceedings - Winter Simulation Conference
Pages934-940
Number of pages7
DOIs
StatePublished - 2006
Event2006 Winter Simulation Conference, WSC - Monterey, CA, United States
Duration: Dec 3 2006Dec 6 2006

Other

Other2006 Winter Simulation Conference, WSC
CountryUnited States
CityMonterey, CA
Period12/3/0612/6/06

Fingerprint

Learning algorithms
Testbeds
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Qiao, H., Rozenblit, J. W., Szidarovszky, F., & Yang, L. (2006). Multi-agent learning model with bargaining. In Proceedings - Winter Simulation Conference (pp. 934-940). [4117702] https://doi.org/10.1109/WSC.2006.323178

Multi-agent learning model with bargaining. / Qiao, Haiyan; Rozenblit, Jerzy W; Szidarovszky, Ferenc; Yang, Lizhi.

Proceedings - Winter Simulation Conference. 2006. p. 934-940 4117702.

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

Qiao, H, Rozenblit, JW, Szidarovszky, F & Yang, L 2006, Multi-agent learning model with bargaining. in Proceedings - Winter Simulation Conference., 4117702, pp. 934-940, 2006 Winter Simulation Conference, WSC, Monterey, CA, United States, 12/3/06. https://doi.org/10.1109/WSC.2006.323178
Qiao H, Rozenblit JW, Szidarovszky F, Yang L. Multi-agent learning model with bargaining. In Proceedings - Winter Simulation Conference. 2006. p. 934-940. 4117702 https://doi.org/10.1109/WSC.2006.323178
Qiao, Haiyan ; Rozenblit, Jerzy W ; Szidarovszky, Ferenc ; Yang, Lizhi. / Multi-agent learning model with bargaining. Proceedings - Winter Simulation Conference. 2006. pp. 934-940
@inproceedings{1e188b6655584b57995965518cb9cbe2,
title = "Multi-agent learning model with bargaining",
abstract = "Decision problems with the features of prisoner's dilemma are quite common. A general solution to this kind of social dilemma is that the agents cooperate to play a joint action. The Nash bargaining solution is an attractive approach to such cooperative games. In this paper, a multi-agent learning algorithm based on the Nash bargaining solution is presented. Different experiments are conducted on a testbed of stochastic games. The experimental results demonstrate that the algorithm converges to the policies of the Nash bargaining solution. Compared with the learning algorithms based on a non-cooperative equilibrium, this algorithm is fast and its complexity is linear with respect to the number of agents and number of iterations. In addition, it avoids the disturbing problem of equilibrium selection.",
author = "Haiyan Qiao and Rozenblit, {Jerzy W} and Ferenc Szidarovszky and Lizhi Yang",
year = "2006",
doi = "10.1109/WSC.2006.323178",
language = "English (US)",
isbn = "1424405017",
pages = "934--940",
booktitle = "Proceedings - Winter Simulation Conference",

}

TY - GEN

T1 - Multi-agent learning model with bargaining

AU - Qiao, Haiyan

AU - Rozenblit, Jerzy W

AU - Szidarovszky, Ferenc

AU - Yang, Lizhi

PY - 2006

Y1 - 2006

N2 - Decision problems with the features of prisoner's dilemma are quite common. A general solution to this kind of social dilemma is that the agents cooperate to play a joint action. The Nash bargaining solution is an attractive approach to such cooperative games. In this paper, a multi-agent learning algorithm based on the Nash bargaining solution is presented. Different experiments are conducted on a testbed of stochastic games. The experimental results demonstrate that the algorithm converges to the policies of the Nash bargaining solution. Compared with the learning algorithms based on a non-cooperative equilibrium, this algorithm is fast and its complexity is linear with respect to the number of agents and number of iterations. In addition, it avoids the disturbing problem of equilibrium selection.

AB - Decision problems with the features of prisoner's dilemma are quite common. A general solution to this kind of social dilemma is that the agents cooperate to play a joint action. The Nash bargaining solution is an attractive approach to such cooperative games. In this paper, a multi-agent learning algorithm based on the Nash bargaining solution is presented. Different experiments are conducted on a testbed of stochastic games. The experimental results demonstrate that the algorithm converges to the policies of the Nash bargaining solution. Compared with the learning algorithms based on a non-cooperative equilibrium, this algorithm is fast and its complexity is linear with respect to the number of agents and number of iterations. In addition, it avoids the disturbing problem of equilibrium selection.

UR - http://www.scopus.com/inward/record.url?scp=46149121782&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46149121782&partnerID=8YFLogxK

U2 - 10.1109/WSC.2006.323178

DO - 10.1109/WSC.2006.323178

M3 - Conference contribution

SN - 1424405017

SN - 9781424405015

SP - 934

EP - 940

BT - Proceedings - Winter Simulation Conference

ER -