Vidal's library
Title: A Distributed Algorithm for Finding Nucleolus-Stable Payoff Divisions
Author: Hrishikesh J. Goradia and José M. Vidal
Book Tittle: Proceedings of the IEEE / WIC / ACM International Conference on Intelligent Agent Technology
Year: 2007
Abstract: The agents in multiagent systems can coordinate their actions and handle tasks jointly by forming coalitions. One of the important steps in this process is the fair division of payoff generated from such a joint effort among all coalition members. The nucleolus is widely recognized as a fair way of distributing the revenue in a coalition. While we have efficient algorithms for computing the nucleolus using linear programming based techniques, we believe that such approaches are infeasible in multiagent settings where the loci of decision making is distributed among all agents in the system, and there is no central agent that can aggregate all data and compute for all agents. Towards this end, in this paper we present a distributed algorithm that computes the nucleolus-stable payoff division for any multiagent system modeled as a characteristic form game. We empirically show that our algorithm has many desirable properties -- it searches only a small fraction of the solution space, evenly distributes the computational load among all agents, and executes reasonably quickly for this hard problem.



@InProceedings{goradia07c,
  author =	 {Hrishikesh J. Goradia and Jos\'{e} M. Vidal},
  title =	 {A Distributed Algorithm for Finding Nucleolus-Stable
                  Payoff Divisions},
  booktitle =	 {Proceedings of the {IEEE} / {WIC} / {ACM}
                  International Conference on Intelligent Agent
                  Technology},
  year =	 2007,
  abstract =	 {The agents in multiagent systems can coordinate
                  their actions and handle tasks jointly by forming
                  coalitions. One of the important steps in this
                  process is the fair division of payoff generated
                  from such a joint effort among all coalition
                  members. The nucleolus is widely recognized as a
                  fair way of distributing the revenue in a
                  coalition. While we have efficient algorithms for
                  computing the nucleolus using linear programming
                  based techniques, we believe that such approaches
                  are infeasible in multiagent settings where the loci
                  of decision making is distributed among all agents
                  in the system, and there is no central agent that
                  can aggregate all data and compute for all
                  agents. Towards this end, in this paper we present a
                  distributed algorithm that computes the
                  nucleolus-stable payoff division for any multiagent
                  system modeled as a characteristic form game. We
                  empirically show that our algorithm has many
                  desirable properties -- it searches only a small
                  fraction of the solution space, evenly distributes
                  the computational load among all agents, and
                  executes reasonably quickly for this hard problem.},
  url = 	 {http://jmvidal.cse.sc.edu/papers/goradia07c.pdf}
}
Last modified: Wed Mar 9 10:16:47 EST 2011