Research Applications of the MAGNET Multi-Agent Contracting Testbed
Abstract
MAGNET is a testbed for exploring decision processes andagent interactions in the domain of multi-agent contracting. We are i nterested in learning how a community of hetero-geneous, self-interested agents, can operate to discover re-sources, make commitments, and carry out plans that in-volvemultipletasksandrequirecoordinationamongagents. We assume that this community of agents contains someagents who have goals that they themselves cannot fully sat-isfy. They may lack the abilities, or they may lack the re-sources, to carry out at least some of the operations in theirplans. Therearealsootherself-interestedagentsinthecom-munity who have resources to offer, and who are willing tomake those resources available to other agents in a way thatmaximizes their value to the agents that control them. The MAGNET system (Collins
Cite
Text
Collins and Gini. "Research Applications of the MAGNET Multi-Agent Contracting Testbed." AAAI Conference on Artificial Intelligence, 2002. doi:10.5555/777092.777266Markdown
[Collins and Gini. "Research Applications of the MAGNET Multi-Agent Contracting Testbed." AAAI Conference on Artificial Intelligence, 2002.](https://mlanthology.org/aaai/2002/collins2002aaai-research/) doi:10.5555/777092.777266BibTeX
@inproceedings{collins2002aaai-research,
title = {{Research Applications of the MAGNET Multi-Agent Contracting Testbed}},
author = {Collins, John and Gini, Maria L.},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2002},
pages = {996-997},
doi = {10.5555/777092.777266},
url = {https://mlanthology.org/aaai/2002/collins2002aaai-research/}
}