Cooperation in Massively Multiagent Systems
Intelligent software agents are computer programs which could interact with each other and are situated in an environment. They are rational and follow their own goals. This could often lead to unwanted situations as the system's goal often does not match the individual goals. Agents want to maximize their own profit. We consider scenarios where a large set of agents exists with interaction links between the agents, which is called a massively multiagent system (MMAS).
Agents have to fulfill jobs which consist of tasks, each requiring a specific skill. The agents differ in their available skills and form a heterogeneous MMAS. Only a fraction of all possible jobs can be done by a single agent. Therefore they need cooperation partners. We do not consider cooperation for free as helping others normally produces costs (i.e. computational power, resources, etc.). The problem of cooperation is therefore a dilemma because helping produces costs which is not consistent to the individual goal of maximizing the profit. But cooperation is the only way to maximize the social welfare, which is not visible to the agents as we only allow them to have limited knowledge.
We want to built up a system of agents producing global cooperation from local interaction and local knowledge.
Contact Person(s)
Publications
Multi-Criteria Cooperation in Multiagent Systems by Local Adaptation.
PhD thesis, University of Paderborn 2011.
[Show BibTeX] [DOI]
@phdthesis{eberling-phd-2011,Determining Cooperation in Multiagent Systems with Cultural Traits.
In Joaquim Filipe and Ana Fred (eds.): Proceedings of the 3rd International Conference on Agents and Artificial Intelligence ICAART 2011 Vol. 2. SciTePress, pp. 173-180, 2011.
[Show BibTeX]
Convergence Analysis of a Multiagent Cooperation Model.
In Joaquim Filipe and Ana Fred (eds.): Proceedings of the 3rd International Conference on Agents and Artificial Intelligence ICAART 2011 Vol. 2. SciTePress, pp. 167-172, 2011.
[Show BibTeX]
Convergence Analysis of a Multiagent Cooperation Model (extended Version).
Technical Report, no. TR-RI-10-321. University of Paderborn 2010.
[Show BibTeX] [PDF]
The Effects of Local Trust Cooperation in Multiagent Systems.
In Jedrzejowicz, Piotr and Nguyen, Ngoc Thanh and Howlett, Robert J. and Jain, Lakhmi C. (eds.): Proceedings of the 4th KES International Symposium on Agent and Multi-Agent Systems: Technologies and Applications. Springer (Heidelberg), Lecture Notes in Computer Science, vol. 6071, pp. 233-242, 2010.
[Show BibTeX] [DOI]
Self-adaptation Strategies to Favor Cooperation.
In Jedrzejowicz, Piotr and Nguyen, Ngoc Thanh and Howlett, Robert J. and Jain, Lakhmi C. (eds.): Proceedings of the 4th KES International Symposium on Agent and Multi-Agent Systems: Technologies and Applications. Springer (Heidelberg), Lecture Notes in Computer Science, vol. 6071, pp. 223-232, 2010.
[Show BibTeX] [DOI]
Towards Determining Cooperation Based on Multiple Criteria.
In Mertsching, Bärbel and Hund, Marcus and Aziz, Muhammad Zaheer (eds.): Proceedings of the 32nd Annual German Conference on AI (KI 2009). Springer, Lecture Notes in Computer Science, vol. 5803, pp. 548-555, 2009.
[Show BibTeX] [DOI]