AAAI Publications, Twenty-Seventh AAAI Conference on Artificial Intelligence

Font Size: 
A Mediation Mechanism for Automated Negotiating Agents Whose Utility Changes over Time
Keisuke Hara, Takayuki Ito

Last modified: 2013-06-29

Abstract


Multi-issue negotiation protocols are an important field of study because real-world negotiation problems are often complex and involve multiple issues. Although much previous work has only addressed linear utility function, that is, simple negotiations involving independent issues, recently, non-linear utility functions for complex negotiations involing interdependent issues have gained attention. Most studies, however, do not focus on the changes in utility space over time. In economic theory, it is often assumed that the utility function changes dynamically over time. It is important to seek the Pareto front, which refers to the set of Pareto optimal points, in negotiation problems. Therefore, in this paper we propose a complex utility space that changes over time and a negotiation mechanism in which the mediator takes the lead in negotiation based on the genetic algorithm (GA). The experimental results show that our approach is suitable for utility that dynamically changes over time, and finds and follows the Pareto front effectively.

Full Text: PDF