Vidal's library
Title: Contract types for satisficing task allocation: II experimental results
Author: Martin R. Andersson and Tuomas Sandholm
Book Tittle: AAAI Spring Symposium: Satisficing Models
Year: 1998
Abstract: We provide experimental results for a task allocation problem where self-interested, individually rational agents (re)contract tasks among themselves. Traditional contract types allow only one task to be transferred between agents at a time (original contracts). In this paper the original and four other contract types are studied: cluster-, swap-, multiagent, and OCSM-contracts. The OCSM-contracts will reach the global optimum even if the agents are individually rational, but in large-scale problems the number of steps required can be prohibitively large, albeit finit. In such cases it is more important to find th best solution reachable in a bounded amount of time. To construct algorithms that achieve that we study different contract types evaluate their performance. This paper dicusses the quality of local optima reached by the different contract types, and how quickly they are reached. It is shown how environmental characteristics such as the number of agents and the number of tasks affects these results. This analysis is used as a basis for making prescriptions about which contract types agents should use in different environments. Out of original-, cluster-, swap-, and multiagent-contracts, either original contracts (if the ration of agents to tasks is great) or cluster-contracts (if the same ratio is small) reach a local optimum with a higher social welfare than the others.

Cited by 12  -  Google Scholar

@InProceedings{andersson98a,
  author =	 {Martin R. Andersson and Tuomas Sandholm},
  title =	 {Contract types for satisficing task allocation: II
                  experimental results},
  booktitle =	 {{AAAI} Spring Symposium: Satisficing Models},
  year =	 1998,
  abstract =	 {We provide experimental results for a task
                  allocation problem where self-interested,
                  individually rational agents (re)contract tasks
                  among themselves. Traditional contract types allow
                  only one task to be transferred between agents at a
                  time (original contracts). In this paper the
                  original and four other contract types are studied:
                  cluster-, swap-, multiagent, and OCSM-contracts. The
                  OCSM-contracts will reach the global optimum even if
                  the agents are individually rational, but in
                  large-scale problems the number of steps required
                  can be prohibitively large, albeit finit. In such
                  cases it is more important to find th best solution
                  reachable in a bounded amount of time. To construct
                  algorithms that achieve that we study different
                  contract types evaluate their performance. This
                  paper dicusses the quality of local optima reached
                  by the different contract types, and how quickly
                  they are reached. It is shown how environmental
                  characteristics such as the number of agents and the
                  number of tasks affects these results. This analysis
                  is used as a basis for making prescriptions about
                  which contract types agents should use in different
                  environments. Out of original-, cluster-, swap-, and
                  multiagent-contracts, either original contracts (if
                  the ration of agents to tasks is great) or
                  cluster-contracts (if the same ratio is small) reach
                  a local optimum with a higher social welfare than
                  the others.},
  keywords =     {multiagent negotiation},
  url =		 {http://jmvidal.cse.sc.edu/library/andersson98a.pdf},
  googleid =	 {oOXmyzf34SAJ:scholar.google.com/},
  citeseer =	 {116375.html},
  cluster = 	 {2369446697989760416}
}
Last modified: Wed Mar 9 10:14:31 EST 2011