How important are updating schemes in multi-agent systems? An illustration on a multi-turmite model.

Nazim Fatès 1 Vincent Chevrier 1
1 MAIA - Autonomous intelligent machine
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : It is to date an open question to know how the updating methods affect the evolution of a multi-agent system. This question has been tackled for various complex systems such as cellular automata, Boolean networks, neural networks but little is known for multi-agent systems, especially for the models with a complex behaviour which emerges from simple local rules. This paper focuses on a multi-turmite model, namely the multiple Langton's ants model. All the agents are updated simultaneously and the variation of the updating scheme consists only in choosing different strate- gies for solving the conflicts produced when two or more agents want to go on the same location. We show that for the same formulation of the agents' behaviour, and the same initial conditions, the use of different updating schemes may lead to qualitatively different evolutions of the system. As a positive spin-off of this study, we exhibit new phenomena of the multi-turmite model such as deadlocks or gliders.
Type de document :
Communication dans un congrès
Ninth International Conference on Autonomous Agents and Multiagent Systems - AAMAS'10, May 2010, Toronto, Canada. ACM, 1, pp.533-540, 2010, Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems. 〈http://www.ifaamas.org/Proceedings/aamas2010/pdf/01%20Full%20Papers/11_04_FP_0210.pdf〉
Liste complète des métadonnées

https://hal.inria.fr/inria-00546845
Contributeur : Nazim Fatès <>
Soumis le : mardi 14 décembre 2010 - 21:33:17
Dernière modification le : jeudi 11 janvier 2018 - 06:19:51

Identifiants

  • HAL Id : inria-00546845, version 1

Collections

Citation

Nazim Fatès, Vincent Chevrier. How important are updating schemes in multi-agent systems? An illustration on a multi-turmite model.. Ninth International Conference on Autonomous Agents and Multiagent Systems - AAMAS'10, May 2010, Toronto, Canada. ACM, 1, pp.533-540, 2010, Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems. 〈http://www.ifaamas.org/Proceedings/aamas2010/pdf/01%20Full%20Papers/11_04_FP_0210.pdf〉. 〈inria-00546845〉

Partager

Métriques

Consultations de la notice

158