Auto-Adaptation of Open MAS through On-Line Modifications of the Environment
This paper proposes a mechanism that is able to encourage agents, participating in an open multiagent system, to follow a desirable behaviour, by introducing modifications in the environment. This mechanism is deployed by using an infrastructure based on institutional agents called incentivators. Incentivators are in charge of discovering which changes of environmental attributes (incentives) have some influence on agents’ preferences. Furthermore, they learn the joint actions that should be promoted in order to improve the system’s utility and applies the identified incentives to induce the desired agent behaviours.
KeywordsIncentives Regulation Adaptation Organisation
- 1.Centeno, R., Billhardt, H.: Using incentive mechanisms for an adaptive regulation of open multi-agent systems. In: Proc. IJCAI 2011 (to appear, 2011)Google Scholar