Vidal's libraryTitle: | Contract types for satisficing task allocation: I theoretical results |
Author: | Tuomas Sandholm |
Book Tittle: | AAAI Spring Symposium: Satisficing Models |
Year: | 1998 |
Abstract: | We analyze task reallocation where individually rational (IR) agents (re)contract tasks among themselves based on marginal costs. A task allocation graph is introduced as a tool for analyzing contract types. Traditional single task contracts always have a short path (sequence of contracts) to the optimal task allocation but an IR path may not exist, or it may not be short. We analyze an algorithm for finding the shortest IR path. Next we introduce cluster contracts, swaps, and multiagent contracts. Each of the four contract types avoids some local optima that the others do not. Even if the protocol is equipped with all four types, local optima exist. To attach this problem, we indroduce OCSM-contract which combine the ideas behind the four earlier types into an atomics contract type. If the protocol is equipped with OCSM-contracts, any sequence of finite number of steps: an oracle--or speculation---is not needed for choosing the path (no subser of OCSM-contracts suffices even with an oracle). This means that the multiagent search does not need to backtrack. This is a powerful result for small problem instances. For large ones, the anytime feature of our multi-contract-type algorithm---with provable monotonic improvement of each agent's solution is more important. |
Cited by 66 - Google Scholar
@InProceedings{sandholm98b,
author = {Tuomas Sandholm},
title = {Contract types for satisficing task allocation: I
theoretical results},
booktitle = {{AAAI} Spring Symposium: Satisficing Models},
year = 1998,
abstract = {We analyze task reallocation where individually
rational (IR) agents (re)contract tasks among
themselves based on marginal costs. A task
allocation graph is introduced as a tool for
analyzing contract types. Traditional single task
contracts always have a short path (sequence of
contracts) to the optimal task allocation but an IR
path may not exist, or it may not be short. We
analyze an algorithm for finding the shortest IR
path. Next we introduce cluster contracts, swaps,
and multiagent contracts. Each of the four contract
types avoids some local optima that the others do
not. Even if the protocol is equipped with all four
types, local optima exist. To attach this problem,
we indroduce OCSM-contract which combine the ideas
behind the four earlier types into an atomics
contract type. If the protocol is equipped with
OCSM-contracts, any sequence of finite number of
steps: an oracle--or speculation---is not needed for
choosing the path (no subser of OCSM-contracts
suffices even with an oracle). This means that the
multiagent search does not need to backtrack. This
is a powerful result for small problem
instances. For large ones, the anytime feature of
our multi-contract-type algorithm---with provable
monotonic improvement of each agent's solution is
more important.},
keywords = {multiagent negotiation},
url = {http://jmvidal.cse.sc.edu/library/sandholm98b.pdf},
googleid = {wJcVeTa89xQJ:scholar.google.com/},
citeseer = {sandholm98contract.html},
cluster = {1510883142151804864}
}
Last modified: Wed Mar 9 10:14:31 EST 2011