Risk-Bounded Formation of Fuzzy Coalitions Among Service Agents
Cooperative autonomous agents form coalitions in order to share and combine resources and services to efficiently respond to market demands. With the variety of resources and services provided online today, there is a need for stable and flexible techniques to support the automation of agent coalition formation in this context. This paper describes an approach to the problem based on fuzzy coalitions. Compared with a classic cooperative game with crisp coalitions (where each agent is a full member of exactly one coalition), an agent can participate in multiple coalitions with varying degrees of involvement. This gives the agents more freedom and flexibility, allowing them to make full use of their resources, thus maximising utility, even if only comparatively small coalitions are formed. An important aspect of our approach is that the agents can control and bound the risk caused by the possible failure or default of some partner agents by spreading their involvement in diverse coalitions.
KeywordsProbability Density Function Membership Degree Coalition Formation Coalition Structure Payoff Distribution
Unable to display preview. Download preview PDF.
- 1.Artzner, P., Delbaen, F., Eber, S., Heath, D.: Coherent measures of risk. Mathematical Finance, 203–228 (1999)Google Scholar
- 3.Blankenburg, B., Klusch, M., Shehory, O.: Fuzzy kernel-stable coalitions between rational agents. In: Proc. 2nd Int. Conference on Autonomous Agents and Multiagent Systems, Melbourne, Australia (2003)Google Scholar
- 4.Bracewell, R.N.: The Fourier Transform and Its Applications, 3rd edn. McGraw-Hill Science/Engineering/Math, New York (1999)Google Scholar
- 6.Chalkiadakis, G., Boutilier, C.: Bayesian reinforcement learning for coalition formation under uncertainty. In: Proc. 3rd Int. Conference on Autonomous Agents and Multiagent Systems, New York, USA (2004)Google Scholar
- 8.Grimmett, G.R., Stirzaker, D.R.: Probability and Random Processes, 3rd edn. Oxford University Press, Oxford (2001)Google Scholar
- 9.Nishizaki, I., Sakawa, M.: Masatoshi Fuzzy and multiobjective games for conflict resolution. Studies in Fuzziness and Soft Computing, vol. 64. Physica-Verlag, Heidelberg (2001)Google Scholar