Computational Intelligence Method for Bargaining Problems with Outside Options
Bargaining is fundamental in social activities. Game-theoretic method is ideal for providing theoretic solutions for simplistic models. Game-theoretic method always requires substantial human efforts in order to solve game-theoretic equilibriums. The analytic complexity increases rapidly when more determinants are considered. We attempt to use an alternative method: computational intelligence, in particular evolutionary algorithms to cope with the increased complexity due to the increase of determinants.
In this paper, we study a relatively complicated bargaining model, in which outside option and discount factor are taken into account. Empirical studies demonstrate that evolutionary algorithms are efficient in finding near-perfect solutions. Experimental results reflect the compound effects of discount factors and outside options upon bargaining outcomes. We argue that evolutionary algorithm is a practically alternative tool for generating reasonably good strategies for fairly complicated bargaining models.
Keywords: Bargaining Problems, Evolutionary Computation
Dr Nanlin Jin
Research Fellow, School of Geography, University of Leeds
My main responsibilities in this project are:
1. Establish social-economic models to simulate involvers’ decision-making;
2. Study farmers’ adaptive learning to dynamic environments;
3. Develop Agent-based model (ABM) system;
4. Integrate the social-economic model with the Biophysics model & Biodiversity model;
5. Support visualization and information retrieval for the Integrated System