A Dependency-Based Automated Negotiation Mechanism for a Hypergraph Utility Model
Akiyuki Mori, Shota Morii, Takayuki Itō · 2015
Interest continues to grow in automated negotiations, especially those that involves multiple interdependent issues, which yield complex nonlinear utility spaces. However, none of the proposed models were able to find a high-quality solution within a realistic time. In this paper, we adopt a dependency-based complex utility model. One of the advanced characteristics of this model is that it can compactly represent the dependencies among issues in the complex utility spaces. Based on this model, we propose a new mediation model that efficiently mediates issue-by-issue negotiations. The mediator sequentially adds negotiation issues based on the dependencies between them for smooth negotiations. The following are the contributions of this paper: By representing the preference profiles of agents based on a hyper graph, we can concisely express the dependencies among issues. Using a mediator that explores based on dependencies, it is possible to extract significant issues for each negotiation participant. As a result, agents can negotiate efficiently in minimum time. We experimentally evaluated our model using several nonlinear utility spaces and showed that it can handle large and complex spaces by finding good solutions.