Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence
Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence
Hi-index | 0.00 |
We investigate the performance of agents co-evolved using genetic programming techniques to play an appropriation common pool game. This game is used to study behaviours of users participating in scenarios with shared resources or interests eg. fisheries. We compare the outcomes achieved by the evolved strategies to that of human players as reported by [6]. Results show that genetic programming techniques are suitable for generating strategies in a repeated investment problem. We find that by using co-evolutionary methods, populations of strategies will quickly converge to nash equilibrium predicted by game theoretic analysis, but also lose many adaptive behaviours. Further, by evolving against a set of naive strategies, we show the creation of diverse and adaptive behaviours that play similarly to humans as described in previous experiments.